<?xml version="1.0" encoding="UTF-8"?>
 <rdf:RDF xmlns="http://purl.org/rss/1.0/" xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:taxo="http://purl.org/rss/1.0/modules/taxonomy/" xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:cc="http://web.resource.org/cc/" xmlns:syn="http://purl.org/rss/1.0/modules/syndication/" xmlns:admin="http://webns.net/mvcb/">
  <channel rdf:about="http://pinboard.in">
    <title>Pinboard (Vaguery)</title>
    <link>https://pinboard.in/u:Vaguery/public/</link>
    <description>recent bookmarks from Vaguery</description>
    <items>
      <rdf:Seq>	<rdf:li rdf:resource="https://arxiv.org/abs/1312.6055v3"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2201.12038"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2401.17720"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2601.21766"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2206.07391"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2407.03357"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2502.08021"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2111.13248"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1802.05448"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2409.05746"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2304.00848"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2303.05228"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2308.04825"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2307.15584"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2302.05116"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2306.15276"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2304.07089"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2105.10541"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1904.04513"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1810.05236"/>
	<rdf:li rdf:resource="https://workfutures.substack.com/p/minimum-viable-work"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1906.04358"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1906.01478"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2012.09715"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1810.07921"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1902.04861"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2012.09318"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2102.05738"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1904.12901"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2012.11780"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2012.11870"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2005.02152"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1704.07067"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1902.07672"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1610.06940"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1904.02063"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1905.03330"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1803.00854"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1904.04412"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1805.01045"/>
	<rdf:li rdf:resource="https://www.pnas.org/content/117/1/43"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2002.08512"/>
	<rdf:li rdf:resource="https://en.wikipedia.org/wiki/RANDU"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1812.07270"/>
	<rdf:li rdf:resource="http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.56.5609"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1812.05163"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1702.08066"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1909.10140"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1802.00296"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1602.04181"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1902.00415"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1909.10140#"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1903.09139"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/math/0402212"/>
	<rdf:li rdf:resource="https://dl.acm.org/citation.cfm?id=2601224"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1708.01559"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1803.02015"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1509.02241"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1608.07617"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1808.09325"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1711.08042"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1806.04236"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1902.11214"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1107.5667"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1807.01672"/>
	<rdf:li rdf:resource="https://ddd.fit.cvut.cz/prj/Benchmarks/index.php?page=download"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1811.01910"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1906.07774"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1802.07481"/>
	<rdf:li rdf:resource="https://hackingsemantics.xyz/2019/leaderboards/"/>
      </rdf:Seq>
    </items>
  </channel><item rdf:about="https://arxiv.org/abs/1312.6055v3">
    <title>[1312.6055v3] Unit Tests for Stochastic Optimization</title>
    <dc:date>2026-07-04T13:14:28+00:00</dc:date>
    <link>https://arxiv.org/abs/1312.6055v3</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Optimization by stochastic gradient descent is an important component of many large-scale machine learning algorithms. A wide variety of such optimization algorithms have been devised; however, it is unclear whether these algorithms are robust and widely applicable across many different optimization landscapes. In this paper we develop a collection of unit tests for stochastic optimization. Each unit test rapidly evaluates an optimization algorithm on a small-scale, isolated, and well-understood difficulty, rather than in real-world scenarios where many such issues are entangled. Passing these unit tests is not sufficient, but absolutely necessary for any algorithms with claims to generality or robustness. We give initial quantitative and qualitative results on numerous established algorithms. The testing framework is open-source, extensible, and easy to apply to new algorithms.
]]></description>
<dc:subject>benchmarking operations-research unit-testing performance-measure rather-interesting metaheuristics neural-networks machine-learning to-write-about to-use</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:ebbe888f0ef2/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:benchmarking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:operations-research"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:unit-testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metaheuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-use"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2201.12038">
    <title>[2201.12038] A survey on flexible/restricted skyline and their applicability</title>
    <dc:date>2026-06-26T12:54:29+00:00</dc:date>
    <link>https://arxiv.org/abs/2201.12038</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Skyline and Top-k are two of the most important methods to extract information from datasets, but both come with their drawbacks, that's why lately some new technics that try to mix the features of the two have been studied. In this survey three new operators are analysed, F-Skyline, ORU/ORD, and ϵ-Skyline. After giving the main ideas behind those and their properties, they are compered on 3 fundamental features such as personalization, cardinality control, and generalization to guide the user to choose the best one for any task.
]]></description>
<dc:subject>multiobjective-optimization software-development-is-not-programming algorithms performance-measure rather-interesting consider:lexicase</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:012c8c11117c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:multiobjective-optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:software-development-is-not-programming"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:lexicase"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2401.17720">
    <title>[2401.17720] Apéry Acceleration of Continued Fractions</title>
    <dc:date>2026-05-25T12:10:58+00:00</dc:date>
    <link>https://arxiv.org/abs/2401.17720</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We explain in detail how to accelerate continued fractions (for constants as well as for functions) using the method used by R.~Apéry in his proof of the irrationality of ζ(3). We show in particular that this can be applied to a large number of continued fractions which can be found in the literature, thus providing a large number of new continued fractions. As examples, we give a new continued fraction for log(2) and for ζ(3), as well as a simple proof of one due to Ramanujan.
]]></description>
<dc:subject>continued-fractions representation rather-interesting heuristics mathematics performance-measure to-write-about to-simulate consider:evolutionary-search consider:accuracy-measures</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:aaaab8b3e49a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:continued-fractions"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:representation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:heuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:mathematics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:evolutionary-search"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:accuracy-measures"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2601.21766">
    <title>[2601.21766] CoFrGeNet: Continued Fraction Architectures for Language Generation</title>
    <dc:date>2026-05-24T17:10:27+00:00</dc:date>
    <link>https://arxiv.org/abs/2601.21766</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Transformers are arguably the preferred architecture for language generation. In this paper, inspired by continued fractions, we introduce a new function class for generative modeling. The architecture family implementing this function class is named CoFrGeNets - Continued Fraction Generative Networks. We design novel architectural components based on this function class that can replace Multi-head Attention and Feed-Forward Networks in Transformer blocks while requiring much fewer parameters. We derive custom gradient formulations to optimize the proposed components more accurately and efficiently than using standard PyTorch-based gradients. Our components are a plug-in replacement requiring little change in training or inference procedures that have already been put in place for Transformer-based models thus making our approach easy to incorporate in large industrial workflows. We experiment on two very different transformer architectures GPT2-xl (1.5B) and Llama3 (3.2B), where the former we pre-train on OpenWebText and GneissWeb, while the latter we pre-train on the docling data mix which consists of nine different datasets. Results show that the performance on downstream classification, Q\& A, reasoning and text understanding tasks of our models is competitive and sometimes even superior to the original models with 23 to 12 the parameters and shorter pre-training time. We believe that future implementations customized to hardware will further bring out the true potential of our architectures.
]]></description>
<dc:subject>machine-learning representation continued-fractions rational-arithmetic rather-interesting neural-networks performance-measure</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:89915d1a6a54/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:representation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:continued-fractions"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rational-arithmetic"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2206.07391">
    <title>[2206.07391] &quot;Why Here and Not There?&quot; -- Diverse Contrasting Explanations of Dimensionality Reduction</title>
    <dc:date>2026-05-24T12:09:35+00:00</dc:date>
    <link>https://arxiv.org/abs/2206.07391</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Dimensionality reduction is a popular preprocessing and a widely used tool in data mining. Transparency, which is usually achieved by means of explanations, is nowadays a widely accepted and crucial requirement of machine learning based systems like classifiers and recommender systems. However, transparency of dimensionality reduction and other data mining tools have not been considered in much depth yet, still it is crucial to understand their behavior -- in particular practitioners might want to understand why a specific sample got mapped to a specific location.
In order to (locally) understand the behavior of a given dimensionality reduction method, we introduce the abstract concept of contrasting explanations for dimensionality reduction, and apply a realization of this concept to the specific application of explaining two dimensional data visualization.
]]></description>
<dc:subject>explanation rather-interesting machine-learning dimension-reduction algorithms performance-measure to-write-about to-do</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:50665668209e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:explanation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:dimension-reduction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-do"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2407.03357">
    <title>[2407.03357] Elementary Formulas for Greatest Common Divisors and Semiprime Factors</title>
    <dc:date>2026-05-24T11:08:57+00:00</dc:date>
    <link>https://arxiv.org/abs/2407.03357</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We conjecture new elementary formulas for computing the greatest common divisor (GCD) of two integers, alongside an elementary formula for extracting the prime factors of semiprimes. These formulas are of fixed-length and require only the basic arithmetic operations of: addition, subtraction, multiplication, division with remainder, and exponentiation. Our GCD formulas result from simplifying a formula of Mazzanti and are derived using Kronecker substitution techniques from our earlier research. By applying these GCD formulas together with our recent discovery of an arithmetic expression for n‾√, we are able to derive explicit elementary formulas for the prime factors of a semiprime n=pq.
]]></description>
<dc:subject>algorithms numerical-methods computational-complexity rather-interesting performance-measure consider:looking-to-see</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:99ab92c256bf/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computational-complexity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:looking-to-see"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2502.08021">
    <title>[2502.08021] Model Selection for Off-policy Evaluation: New Algorithms and Experimental Protocol</title>
    <dc:date>2025-04-16T22:26:45+00:00</dc:date>
    <link>https://arxiv.org/abs/2502.08021</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Holdout validation and hyperparameter tuning from data is a long-standing problem in offline reinforcement learning (RL). A standard framework is to use off-policy evaluation (OPE) methods to evaluate and select the policies, but OPE either incurs exponential variance (e.g., importance sampling) or has hyperparameters on their own (e.g., FQE and model-based). In this work we focus on hyperparameter tuning for OPE itself, which is even more under-investigated. Concretely, we select among candidate value functions ("model-free") or dynamics ("model-based") to best assess the performance of a target policy. Our contributions are two fold. We develop: (1) new model-free and model-based selectors with theoretical guarantees, and (2) a new experimental protocol for empirically evaluating them. Compared to the model-free protocol in prior works, our new protocol allows for more stable generation of candidate value functions, better control of misspecification, and evaluation of model-free and model-based methods alike. We exemplify the protocol on a Gym environment, and find that our new model-free selector, LSTD-Tournament, demonstrates promising empirical performance.
]]></description>
<dc:subject>machine-learning validation training-vs-testing to-understand performance-measure hyperparameter-tuning</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:7fe943a0ebd9/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:validation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:training-vs-testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:hyperparameter-tuning"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2111.13248">
    <title>[2111.13248] Heuristic Search of (Semi-)Bent Functions based on Cellular Automata</title>
    <dc:date>2024-10-08T19:53:09+00:00</dc:date>
    <link>https://arxiv.org/abs/2111.13248</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[An interesting thread in the research of Boolean functions for cryptography and coding theory is the study of secondary constructions: given a known function with a good cryptographic profile, the aim is to extend it to a (usually larger) function possessing analogous properties. In this work, we continue the investigation of a secondary construction based on cellular automata, focusing on the classes of bent and semi-bent functions. We prove that our construction preserves the algebraic degree of the local rule, and we narrow our attention to the subclass of quadratic functions, performing several experiments based on exhaustive combinatorial search and heuristic optimization through Evolutionary Strategies (ES). Finally, we classify the obtained results up to permutation equivalence, remarking that the number of equivalence classes that our CA-XOR construction can successfully extend grows very quickly with respect to the CA diameter.
]]></description>
<dc:subject>cryptography cellular-automata performance-measure evolutionary-algorithms rather-interesting to-understand to-simulate consider:breadth</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:5b8eab373cfe/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:cryptography"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:cellular-automata"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:evolutionary-algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:breadth"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1802.05448">
    <title>[1802.05448] Discrepancy-based Evolutionary Diversity Optimization</title>
    <dc:date>2024-10-06T21:35:51+00:00</dc:date>
    <link>https://arxiv.org/abs/1802.05448</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Diversity plays a crucial role in evolutionary computation. While diversity has been mainly used to prevent the population of an evolutionary algorithm from premature convergence, the use of evolutionary algorithms to obtain a diverse set of solutions has gained increasing attention in recent years. Diversity optimization in terms of features on the underlying problem allows to obtain a better understanding of possible solutions to the problem at hand and can be used for algorithm selection when dealing with combinatorial optimization problems such as the Traveling Salesperson Problem. We explore the use of the star-discrepancy measure to guide the diversity optimization process of an evolutionary algorithm.
In our experimental investigations, we consider our discrepancy-based diversity optimization approaches for evolving diverse sets of images as well as instances of the Traveling Salesperson problem where a local search is not able to find near optimal solutions. Our experimental investigations comparing three diversity optimization approaches show that a discrepancy-based diversity optimization approach using a tie-breaking rule based on weighted differences to surrounding feature points provides the best results in terms of the star discrepancy measure.
]]></description>
<dc:subject>evolutionary-algorithms discrepancy performance-measure rather-interesting to-understand consider:lexicase consider:representation</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:351b253bf73a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:evolutionary-algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:discrepancy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:lexicase"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:representation"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2409.05746">
    <title>[2409.05746] LLMs Will Always Hallucinate, and We Need to Live With This</title>
    <dc:date>2024-09-19T15:37:19+00:00</dc:date>
    <link>https://arxiv.org/abs/2409.05746</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[As Large Language Models become more ubiquitous across domains, it becomes important to examine their inherent limitations critically. This work argues that hallucinations in language models are not just occasional errors but an inevitable feature of these systems. We demonstrate that hallucinations stem from the fundamental mathematical and logical structure of LLMs. It is, therefore, impossible to eliminate them through architectural improvements, dataset enhancements, or fact-checking mechanisms. Our analysis draws on computational theory and Godel's First Incompleteness Theorem, which references the undecidability of problems like the Halting, Emptiness, and Acceptance Problems. We demonstrate that every stage of the LLM process-from training data compilation to fact retrieval, intent classification, and text generation-will have a non-zero probability of producing hallucinations. This work introduces the concept of Structural Hallucination as an intrinsic nature of these systems. By establishing the mathematical certainty of hallucinations, we challenge the prevailing notion that they can be fully mitigated.
]]></description>
<dc:subject>machine-learning neural-networks hallucination performance-measure rather-interesting statistics philosophy-of-engineering</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:ec90a32de8c2/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:hallucination"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:philosophy-of-engineering"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2304.00848">
    <title>[2304.00848] Towards Goal-Oriented Semantic Communications: New Metrics, Open Challenges, and Future Research Directions</title>
    <dc:date>2024-08-08T14:02:16+00:00</dc:date>
    <link>https://arxiv.org/abs/2304.00848</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Since Shannon's pioneering masterpiece which established the prototype of modern information theory, the goal of communications has long been promising accurate message reconstruction at a refined bit-by-bit level, which deliberately neglects the semantics/effectiveness aspect of information. Nevertheless, the recent development of wireless technologies and the spurt of deep learning (DL) techniques allow us to reclaim the meaning/usefulness aspect in the design of future 6G communication paradigms, where goal-oriented communication is becoming a trend. Age of Information (AoI), a well-known metric that captures the importance of information by recording the time elapsed from the generation time slot, has been extended to various variants, such as Value of Information (VoI), Urgency of Information (UoI), Age of Incorrect Information (AoII), and etc. While each of them proposes novel ways to measure the meaning/usefulness aspect of information, there is not yet an integrated framework encompassing all of them. To this end, we propose a novel tensor-based approach, the Goal-oriented Tensor (GoT), to unify them, which also allows more flexible and fine-grained goal characterizations. Following the proposed GoT, we architect a holistic goal-oriented framework to enable goal-oriented semantic communications, in which information perception, dissemination, and control-plane decisions are implemented in accordance with specific goals. Finally, we outline several open challenges to fulfill the vision of the GoT framework.
]]></description>
<dc:subject>define-your-terms information-theory semantics huh seems-a-bit-what's-the-word to-understand performance-measure metrics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:813abfc43434/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:define-your-terms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:information-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:semantics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:huh"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:seems-a-bit-what's-the-word"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metrics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2303.05228">
    <title>[2303.05228] A classification of S-boxes generated by Orthogonal Cellular Automata</title>
    <dc:date>2023-09-09T13:24:00+00:00</dc:date>
    <link>https://arxiv.org/abs/2303.05228</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Most of the approaches published in the literature to construct S-boxes via Cellular Automata (CA) work by either iterating a finite CA for several time steps, or by a one-shot application of the global rule. The main characteristic that brings together these works is that they employ a single CA rule to define the vectorial Boolean function of the S-box. In this work, we explore a different direction for the design of S-boxes that leverages on Orthogonal CA (OCA), i.e. pairs of CA rules giving rise to orthogonal Latin squares. The motivation stands on the facts that an OCA pair already defines a bijective transformation, and moreover the orthogonality property of the resulting Latin squares ensures a minimum amount of diffusion. We exhaustively enumerate all S-boxes generated by OCA pairs of diameter 4≤d≤6, and measure their nonlinearity. Interestingly, we observe that for d=4 and d=5 all S-boxes are linear, despite the underlying CA local rules being nonlinear. The smallest nonlinear S-boxes emerges for d=6, but their nonlinearity is still too low to be used in practice. Nonetheless, we unearth an interesting structure of linear OCA S-boxes, proving that their Linear Components Space (LCS) is itself the image of a linear CA, or equivalently a polynomial code. We finally classify all linear OCA S-boxes in terms of their generator polynomials.
]]></description>
<dc:subject>cellular-automata performance-measure cryptography to-understand rather-interesting consider:low-discrepancy-values consider:visualization</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:d75e6f11eb30/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:cellular-automata"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:cryptography"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:low-discrepancy-values"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:visualization"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2308.04825">
    <title>[2308.04825] Repelled point processes with application to numerical integration</title>
    <dc:date>2023-09-09T13:01:14+00:00</dc:date>
    <link>https://arxiv.org/abs/2308.04825</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Linear statistics of point processes yield Monte Carlo estimators of integrals. While the simplest approach relies on a homogeneous Poisson point process, more regularly spread point processes, such as scrambled low-discrepancy sequences or determinantal point processes, can yield Monte Carlo estimators with fast-decaying mean square error. Following the intuition that more regular configurations result in lower integration error, we introduce the repulsion operator, which reduces clustering by slightly pushing the points of a configuration away from each other. Our main theoretical result is that applying the repulsion operator to a homogeneous Poisson point process yields an unbiased Monte Carlo estimator with lower variance than under the original point process. On the computational side, the evaluation of our estimator is only quadratic in the number of integrand evaluations and can be easily parallelized without any communication across tasks. We illustrate our variance reduction result with numerical experiments and compare it to popular Monte Carlo methods. Finally, we numerically investigate a few open questions on the repulsion operator. In particular, the experiments suggest that the variance reduction also holds when the operator is applied to other motion-invariant point processes.
]]></description>
<dc:subject>low-discrepancy-sequence numerical-methods sampling algorithms horse-races rather-interesting approximation performance-measure nonlinear-dynamics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:1d714b506f44/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:low-discrepancy-sequence"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:horse-races"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:approximation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:nonlinear-dynamics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2307.15584">
    <title>[2307.15584] Quasi-Monte Carlo Algorithms (not only) for Graphics Software</title>
    <dc:date>2023-09-09T12:58:09+00:00</dc:date>
    <link>https://arxiv.org/abs/2307.15584</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Quasi-Monte Carlo methods have become the industry standard in computer graphics. For that purpose, efficient algorithms for low discrepancy sequences are discussed. In addition, numerical pitfalls encountered in practice are revealed. We then take a look at massively parallel quasi-Monte Carlo integro-approximation for image synthesis by light transport simulation. Beyond superior uniformity, low discrepancy points may be optimized with respect to additional criteria, such as noise characteristics at low sampling rates or the quality of low-dimensional projections.
]]></description>
<dc:subject>low-discrepancy-numbers algorithms numerical-methods computer-graphics sampling rather-interesting performance-measure to-write-about to-cite</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:3865da3252e2/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:low-discrepancy-numbers"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computer-graphics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-cite"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2302.05116">
    <title>[2302.05116] Example-Based Sampling with Diffusion Models</title>
    <dc:date>2023-09-09T12:49:21+00:00</dc:date>
    <link>https://arxiv.org/abs/2302.05116</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Much effort has been put into developing samplers with specific properties, such as producing blue noise, low-discrepancy, lattice or Poisson disk samples. These samplers can be slow if they rely on optimization processes, may rely on a wide range of numerical methods, are not always differentiable. The success of recent diffusion models for image generation suggests that these models could be appropriate for learning how to generate point sets from examples. However, their convolutional nature makes these methods impractical for dealing with scattered data such as point sets. We propose a generic way to produce 2-d point sets imitating existing samplers from observed point sets using a diffusion model. We address the problem of convolutional layers by leveraging neighborhood information from an optimal transport matching to a uniform grid, that allows us to benefit from fast convolutions on grids, and to support the example-based learning of non-uniform sampling patterns. We demonstrate how the differentiability of our approach can be used to optimize point sets to enforce properties.
]]></description>
<dc:subject>numerical-methods low-discrepancy-numbers quasirandom-numbers algorithms rather-interesting sampling performance-measure to-understand to-cite</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:e86eac84f2cb/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:low-discrepancy-numbers"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:quasirandom-numbers"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-cite"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2306.15276">
    <title>[2306.15276] Heuristic Approaches to Obtain Low-Discrepancy Point Sets via Subset Selection</title>
    <dc:date>2023-08-13T11:20:23+00:00</dc:date>
    <link>https://arxiv.org/abs/2306.15276</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Building upon the exact methods presented in our earlier work [J. Complexity, 2022], we introduce a heuristic approach for the star discrepancy subset selection problem. The heuristic gradually improves the current-best subset by replacing one of its elements at a time. While we prove that the heuristic does not necessarily return an optimal solution, we obtain very promising results for all tested dimensions. For example, for moderate point set sizes 30≤n≤240 in dimension 6, we obtain point sets with L∞ star discrepancy up to 35% better than that of the first n points of the Sobol' sequence. Our heuristic works in all dimensions, the main limitation being the precision of the discrepancy calculation algorithms. 
We also provide a comparison with a recent energy functional introduced by Steinerberger [J. Complexity, 2019], showing that our heuristic performs better on all tested instances.
]]></description>
<dc:subject>low-discrepancy performance-measure heuristics sampling rather-interesting looking-to-see generative-models to-write-about to-simulate consider:genetic-programming</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:ce011607b322/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:low-discrepancy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:heuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:generative-models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:genetic-programming"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2304.07089">
    <title>[2304.07089] Analyzing the Interaction Between Down-Sampling and Selection</title>
    <dc:date>2023-08-13T11:12:48+00:00</dc:date>
    <link>https://arxiv.org/abs/2304.07089</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Genetic programming systems often use large training sets to evaluate the quality of candidate solutions for selection. However, evaluating populations on large training sets can be computationally expensive. Down-sampling training sets has long been used to decrease the computational cost of evaluation in a wide range of application domains. Indeed, recent studies have shown that both random and informed down-sampling can substantially improve problem-solving success for GP systems that use the lexicase parent selection algorithm. We use the PushGP framework to experimentally test whether these down-sampling techniques can also improve problem-solving success in the context of two other commonly used selection methods, fitness-proportionate and tournament selection, across eight GP problems (four program synthesis and four symbolic regression). We verified that down-sampling can benefit the problem-solving success of both fitness-proportionate and tournament selection. However, the number of problems wherein down-sampling improved problem-solving success varied by selection scheme, suggesting that the impact of down-sampling depends both on the problem and choice of selection scheme. Surprisingly, we found that down-sampling was most consistently beneficial when combined with lexicase selection as compared to tournament and fitness-proportionate selection. Overall, our results suggest that down-sampling should be considered more often when solving test-based GP problems.
]]></description>
<dc:subject>lexicase genetic-programming sampling metaheuristics performance-measure hey-I-know-this-guy to-write-about consider:counting-solutions consider:landscape consider:multiobjective-tradeoffs</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:70902eeeb474/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:lexicase"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:genetic-programming"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metaheuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:hey-I-know-this-guy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:counting-solutions"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:landscape"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:multiobjective-tradeoffs"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2105.10541">
    <title>[2105.10541] Addressing the Multiplicity of Solutions in Optical Lens Design as a Niching Evolutionary Algorithms Computational Challenge</title>
    <dc:date>2023-02-07T12:58:57+00:00</dc:date>
    <link>https://arxiv.org/abs/2105.10541</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Optimal Lens Design constitutes a fundamental, long-standing real-world optimization challenge. Potentially large number of optima, rich variety of critical points, as well as solid understanding of certain optimal designs per simple problem instances, provide altogether the motivation to address it as a niching challenge. This study applies established Niching-CMA-ES heuristic to tackle this design problem (6-dimensional Cooke triplet) in a simulation-based fashion. The outcome of employing Niching-CMA-ES `out-of-the-box' proves successful, and yet it performs best when assisted by a local searcher which accurately drives the search into optima. The obtained search-points are corroborated based upon concrete knowledge of this problem-instance, accompanied by gradient and Hessian calculations for validation. We extensively report on this computational campaign, which overall resulted in (i) the location of 19 out of 21 known minima within a single run, (ii) the discovery of 540 new optima. These are new minima similar in shape to 21 theoretical solutions, but some of them have better merit function value (unknown heretofore), (iii) the identification of numerous infeasibility pockets throughout the domain (also unknown heretofore). We conclude that niching mechanism is well-suited to address this problem domain, and hypothesize on the apparent multidimensional structures formed by the attained new solutions.
]]></description>
<dc:subject>evolutionary-algorithms multiobjective-optimization machine-learning engineering-design rather-interesting performance-measure to-simulate consider:visualization optics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:461cdf4aac8d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:evolutionary-algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:multiobjective-optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:visualization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1904.04513">
    <title>[1904.04513] Towards a complete perspective on labeled tree indexing: new size bounds, efficient constructions, and beyond</title>
    <dc:date>2022-03-04T11:27:07+00:00</dc:date>
    <link>https://arxiv.org/abs/1904.04513</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[A labeled tree (or a trie) is a natural generalization of a string, which can also be seen as a compact representation of a set of strings. This paper considers the labeled tree indexing problem, and provides a number of new results on space bound analysis, and on algorithms for efficient construction and pattern matching queries. Kosaraju [FOCS 1989] was the first to consider the labeled tree indexing problem, and he proposed the suffix tree for a backward trie, where the strings in the trie are read in the leaf-to-root direction. In contrast to a backward trie, we call a usual trie as a forward trie. Despite a few follow-up works after Kosaraju's paper, indexing forward/backward tries is not well understood yet. In this paper, we show a full perspective on the sizes of indexing structures such as suffix trees, DAWGs, CDAWGs, suffix arrays, affix trees, affix arrays for forward and backward tries. Some of them take O(n) space in the size n of the input trie, while the others can occupy O(n2) space in the worst case. In particular, we show that the size of the DAWG for a forward trie with n nodes is Ω(σn), where σ is the number of distinct characters in the trie. This becomes Ω(n2) for an alphabet of size σ=Θ(n). Still, we show that there is a compact O(n)-space implicit representation of the DAWG for a forward trie, whose space requirement is independent of the alphabet size. This compact representation allows for simulating each DAWG edge traversal in O(logσ) time, and can be constructed in O(n) time and space over any integer alphabet of size O(n). In addition, this readily extends to the first indexing structure that permits bidirectional pattern searches over a trie within linear space in the input trie size.
]]></description>
<dc:subject>strings data-structures optimization algorithms rather-interesting performance-measure information-theory engineering-design to-write-about to-simulate consider:error</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:7cd802a335d2/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:strings"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:data-structures"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:information-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:error"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1810.05236">
    <title>[1810.05236] Practical Design Space Exploration</title>
    <dc:date>2021-08-07T11:39:21+00:00</dc:date>
    <link>https://arxiv.org/abs/1810.05236</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Multi-objective optimization is a crucial matter in computer systems design space exploration because real-world applications often rely on a trade-off between several objectives. Derivatives are usually not available or impractical to compute and the feasibility of an experiment can not always be determined in advance. These problems are particularly difficult when the feasible region is relatively small, and it may be prohibitive to even find a feasible experiment, let alone an optimal one. 
We introduce a new methodology and corresponding software framework, HyperMapper 2.0, which handles multi-objective optimization, unknown feasibility constraints, and categorical/ordinal variables. This new methodology also supports injection of the user prior knowledge in the search when available. All of these features are common requirements in computer systems but rarely exposed in existing design space exploration systems. The proposed methodology follows a white-box model which is simple to understand and interpret (unlike, for example, neural networks) and can be used by the user to better understand the results of the automatic search. 
We apply and evaluate the new methodology to the automatic static tuning of hardware accelerators within the recently introduced Spatial programming language, with minimization of design run-time and compute logic under the constraint of the design fitting in a target field-programmable gate array chip. Our results show that HyperMapper 2.0 provides better Pareto fronts compared to state-of-the-art baselines, with better or competitive hypervolume indicator and with 8x improvement in sampling budget for most of the benchmarks explored.
]]></description>
<dc:subject>multiobjective-optimization exploration-and-exploitation rather-interesting algorithms metaheuristics performance-measure to-write-about to-understand</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:cd49c7035aac/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:multiobjective-optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:exploration-and-exploitation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metaheuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://workfutures.substack.com/p/minimum-viable-work">
    <title>Minimum Viable Work - work futures</title>
    <dc:date>2021-07-24T11:58:51+00:00</dc:date>
    <link>https://workfutures.substack.com/p/minimum-viable-work</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[I am now fascinated with a fork off of Sahil’s insight, one that I will be pursuing in the coming months: Minimum Viable Work. How can companies operate with the greatest degree of individual autonomy, the lowest degree of managerial overhead, and the highest levels of cooperation without coercion?

]]></description>
<dc:subject>worklife social-norms rather-interesting working-from-home performance-measure institutional-design</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:3174ea5ebba4/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:worklife"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:social-norms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:working-from-home"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:institutional-design"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1906.04358">
    <title>[1906.04358] Weight Agnostic Neural Networks</title>
    <dc:date>2021-07-16T10:24:50+00:00</dc:date>
    <link>https://arxiv.org/abs/1906.04358</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Not all neural network architectures are created equal, some perform much better than others for certain tasks. But how important are the weight parameters of a neural network compared to its architecture? In this work, we question to what extent neural network architectures alone, without learning any weight parameters, can encode solutions for a given task. We propose a search method for neural network architectures that can already perform a task without any explicit weight training. To evaluate these networks, we populate the connections with a single shared weight parameter sampled from a uniform random distribution, and measure the expected performance. We demonstrate that our method can find minimal neural network architectures that can perform several reinforcement learning tasks without weight training. On a supervised learning domain, we find network architectures that achieve much higher than chance accuracy on MNIST using random weights. Interactive version of this paper at this https URL
]]></description>
<dc:subject>neural-networks machine-learning performance-measure rather-interesting implicit-objective architecture m algorithms</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:d2874aa61939/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:implicit-objective"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:architecture"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:m"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1906.01478">
    <title>[1906.01478] What do AI algorithms actually learn? - On false structures in deep learning</title>
    <dc:date>2021-07-12T23:51:46+00:00</dc:date>
    <link>https://arxiv.org/abs/1906.01478</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[There are two big unsolved mathematical questions in artificial intelligence (AI): (1) Why is deep learning so successful in classification problems and (2) why are neural nets based on deep learning at the same time universally unstable, where the instabilities make the networks vulnerable to adversarial attacks. We present a solution to these questions that can be summed up in two words; false structures. Indeed, deep learning does not learn the original structures that humans use when recognising images (cats have whiskers, paws, fur, pointy ears, etc), but rather different false structures that correlate with the original structure and hence yield the success. However, the false structure, unlike the original structure, is unstable. The false structure is simpler than the original structure, hence easier to learn with less data and the numerical algorithm used in the training will more easily converge to the neural network that captures the false structure. We formally define the concept of false structures and formulate the solution as a conjecture. Given that trained neural networks always are computed with approximations, this conjecture can only be established through a combination of theoretical and computational results similar to how one establishes a postulate in theoretical physics (e.g. the speed of light is constant). Establishing the conjecture fully will require a vast research program characterising the false structures. We provide the foundations for such a program establishing the existence of the false structures in practice. Finally, we discuss the far reaching consequences the existence of the false structures has on state-of-the-art AI and Smale's 18th problem.
]]></description>
<dc:subject>machine-learning ontology philosophy-of-engineering performance-measure internal-state rather-interesting to-write-about experiment</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:83f66d016382/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:ontology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:philosophy-of-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:internal-state"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:experiment"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2012.09715">
    <title>[2012.09715] Approximating inverse cumulative distribution functions to produce approximate random variables</title>
    <dc:date>2021-06-27T10:12:23+00:00</dc:date>
    <link>https://arxiv.org/abs/2012.09715</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[For random variables produced through the inverse transform method, approximate random variables are introduced, which are produced by approximations to a distribution's inverse cumulative distribution function. These approximations are designed to be computationally inexpensive, and much cheaper than exact library functions, and thus highly suitable for use in Monte Carlo simulations. Two approximations are presented for the Gaussian distribution: a piecewise constant on equally spaced intervals, and a piecewise linear using geometrically decaying intervals. The error of the approximations are bounded and the convergence demonstrated, and the computational savings measured for C and C++ implementations. Implementations tailored for Intel and Arm hardwares are inspected, alongside hardware agnostic implementations built using OpenMP. The savings are incorporated into a nested multilevel Monte Carlo framework with the Euler-Maruyama scheme to exploit the speed ups without losing accuracy, offering speed ups by a factor of 5--7. These ideas are empirically extended to the Milstein scheme, and the Cox-Ingersoll-Ross process' non central chi-squared distribution, which offer speed ups by a factor of 250 or more.
]]></description>
<dc:subject>numerical-methods approximation rather-interesting simulation sampling performance-measure computational-complexity nudge-targets consider:representation consider:performance-measures consider:lexicase consider:extreme-values</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:8ed61cde1044/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:approximation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:simulation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computational-complexity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:nudge-targets"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:representation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:performance-measures"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:lexicase"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:extreme-values"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1810.07921">
    <title>[1810.07921] Concentration of the Frobenius norm of generalized matrix inverses</title>
    <dc:date>2021-05-23T21:40:02+00:00</dc:date>
    <link>https://arxiv.org/abs/1810.07921</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In many applications it is useful to replace the Moore-Penrose pseudoinverse (MPP) by a different generalized inverse with more favorable properties. We may want, for example, to have many zero entries, but without giving up too much of the stability of the MPP. One way to quantify stability is by how much the Frobenius norm of a generalized inverse exceeds that of the MPP. In this paper we derive finite-size concentration bounds for the Frobenius norm of ℓp-minimal general inverses of iid Gaussian matrices, with 1≤p≤2. For p=1 we prove exponential concentration of the Frobenius norm of the sparse pseudoinverse; for p=2, we get a similar concentration bound for the MPP. Our proof is based on the convex Gaussian min-max theorem, but unlike previous applications which give asymptotic results, we derive finite-size bounds.
]]></description>
<dc:subject>matrices generalization rather-interesting inverse-problems performance-measure to-write-about consider:hillclimbing consider:diversity-of-sampling</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:b99b9dd7796c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:matrices"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:generalization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:inverse-problems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:hillclimbing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:diversity-of-sampling"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1902.04861">
    <title>[1902.04861] Which Neural Network Architecture matches Human Behavior in Artificial Grammar Learning?</title>
    <dc:date>2021-05-23T11:34:59+00:00</dc:date>
    <link>https://arxiv.org/abs/1902.04861</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In recent years artificial neural networks achieved performance close to or better than humans in several domains: tasks that were previously human prerogatives, such as language processing, have witnessed remarkable improvements in state of the art models. One advantage of this technological boost is to facilitate comparison between different neural networks and human performance, in order to deepen our understanding of human cognition. Here, we investigate which neural network architecture (feed-forward vs. recurrent) matches human behavior in artificial grammar learning, a crucial aspect of language acquisition. Prior experimental studies proved that artificial grammars can be learnt by human subjects after little exposure and often without explicit knowledge of the underlying rules. We tested four grammars with different complexity levels both in humans and in feedforward and recurrent networks. Our results show that both architectures can 'learn' (via error back-propagation) the grammars after the same number of training sequences as humans do, but recurrent networks perform closer to humans than feedforward ones, irrespective of the grammar complexity level. Moreover, similar to visual processing, in which feedforward and recurrent architectures have been related to unconscious and conscious processes, our results suggest that explicit learning is best modeled by recurrent architectures, whereas feedforward networks better capture the dynamics involved in implicit learning.
]]></description>
<dc:subject>neural-networks out-of-the-box rather-interesting reverse-engineering off-label-modeling theoretical-biology performance-measure to-write-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:3c9735eb8992/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:out-of-the-box"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:reverse-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:off-label-modeling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:theoretical-biology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2012.09318">
    <title>[2012.09318] Applying Deutsch's concept of good explanations to artificial intelligence and neuroscience -- an initial exploration</title>
    <dc:date>2021-05-22T22:39:16+00:00</dc:date>
    <link>https://arxiv.org/abs/2012.09318</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Artificial intelligence has made great strides since the deep learning revolution, but AI systems still struggle to extrapolate outside of their training data and adapt to new situations. For inspiration we look to the domain of science, where scientists have been able to develop theories which show remarkable ability to extrapolate and sometimes predict the existence of phenomena which have never been observed before. According to David Deutsch, this type of extrapolation, which he calls "reach", is due to scientific theories being hard to vary. In this work we investigate Deutsch's hard-to-vary principle and how it relates to more formalized principles in deep learning such as the bias-variance trade-off and Occam's razor. We distinguish internal variability, how much a model/theory can be varied internally while still yielding the same predictions, with external variability, which is how much a model must be varied to accurately predict new, out-of-distribution data. We discuss how to measure internal variability using the size of the Rashomon set and how to measure external variability using Kolmogorov complexity. We explore what role hard-to-vary explanations play in intelligence by looking at the human brain and distinguish two learning systems in the brain. The first system operates similar to deep learning and likely underlies most of perception and motor control while the second is a more creative system capable of generating hard-to-vary explanations of the world. We argue that figuring out how replicate this second system, which is capable of generating hard-to-vary explanations, is a key challenge which needs to be solved in order to realize artificial general intelligence. We make contact with the framework of Popperian epistemology which rejects induction and asserts that knowledge generation is an evolutionary process which proceeds through conjecture and refutation.
]]></description>
<dc:subject>artificial-intelligence performance-measure philosophy-of-science philosophy-of-engineering knows-it-when-ya-sees-it</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:de20f35dc8ef/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:artificial-intelligence"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:philosophy-of-science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:philosophy-of-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:knows-it-when-ya-sees-it"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2102.05738">
    <title>[2102.05738] Refinement of polygonal grids using Convolutional Neural Networks with applications to polygonal Discontinous Galerkin and Virtual Element methods</title>
    <dc:date>2021-05-19T10:55:00+00:00</dc:date>
    <link>https://arxiv.org/abs/2102.05738</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We propose new strategies to handle polygonal grids refinement based on Convolutional Neural Networks (CNNs). We show that CNNs can be successfully employed to identify correctly the "shape" of a polygonal element so as to design suitable refinement criteria to be possibly employed within adaptive refinement strategies. We propose two refinement strategies that exploit the use of CNNs to classify elements' shape, at a low computational cost. We test the proposed idea considering two families of finite element methods that support arbitrarily shaped polygonal elements, namely Polygonal Discontinuous Galerkin (PolyDG) methods and Virtual Element Methods (VEMs). We demonstrate that the proposed algorithms can greatly improve the performance of the discretization schemes both in terms of accuracy and quality of the underlying grids. Moreover, since the training phase is performed off-line and is problem independent the overall computational costs are kept low.
]]></description>
<dc:subject>neural-networks lattices graph-theory performance-measure plane-geometry to-understand consider:performance-measures</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:4d9cc5f1474e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:lattices"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:graph-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:plane-geometry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:performance-measures"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1904.12901">
    <title>[1904.12901] Challenges of Real-World Reinforcement Learning</title>
    <dc:date>2021-05-19T10:48:38+00:00</dc:date>
    <link>https://arxiv.org/abs/1904.12901</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Reinforcement learning (RL) has proven its worth in a series of artificial domains, and is beginning to show some successes in real-world scenarios. However, much of the research advances in RL are often hard to leverage in real-world systems due to a series of assumptions that are rarely satisfied in practice. We present a set of nine unique challenges that must be addressed to productionize RL to real world problems. For each of these challenges, we specify the exact meaning of the challenge, present some approaches from the literature, and specify some metrics for evaluating that challenge. An approach that addresses all nine challenges would be applicable to a large number of real world problems. We also present an example domain that has been modified to present these challenges as a testbed for practical RL research.
]]></description>
<dc:subject>machine-learning reinforcement-learning algorithms theory-and-practice-sitting-in-a-tree rather-interesting pragmatics to-write-about consider:robustness consider:noise performance-measure</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:a457ad6632d4/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:reinforcement-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:theory-and-practice-sitting-in-a-tree"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:pragmatics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:robustness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:noise"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2012.11780">
    <title>[2012.11780] Towards an Automatic System for Extracting Planar Orientations from Software Generated Point Clouds</title>
    <dc:date>2021-05-09T11:42:24+00:00</dc:date>
    <link>https://arxiv.org/abs/2012.11780</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In geology, a key activity is the characterisation of geological structures (surface formation topology and rock units) using Planar Orientation measurements such as Strike, Dip and Dip Direction. In general these measurements are collected manually using basic equipment; usually a compass/clinometer and a backboard, recorded on a map by hand. Various computing techniques and technologies, such as Lidar, have been utilised in order to automate this process and update the collection paradigm for these types of measurements. Techniques such as Structure from Motion (SfM) reconstruct of scenes and objects by generating a point cloud from input images, with detailed reconstruction possible on the decimetre scale. SfM-type techniques provide advantages in areas of cost and usability in more varied environmental conditions, while sacrificing the extreme levels of data fidelity. Here is presented a methodology of data acquisition and a Machine Learning-based software system: GeoStructure, developed to automate the measurement of orientation measurements. Rather than deriving measurements using a method applied to the input images, such as the Hough Transform, this method takes measurements directly from the reconstructed point cloud surfaces. Point cloud noise is mitigated using a Mahalanobis distance implementation. Significant structure is characterised using a k-nearest neighbour region growing algorithm, and final surface orientations are quantified using the plane, and normal direction cosines.
]]></description>
<dc:subject>machine-learning inference data-analysis geology image-processing rather-interesting constraint-satisfaction performance-measure</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:8a8c856e99b0/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:data-analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:geology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:image-processing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:constraint-satisfaction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2012.11870">
    <title>[2012.11870] A Survey of Methods for Managing the Classification and Solution of Data Imbalance Problem</title>
    <dc:date>2021-05-09T11:40:02+00:00</dc:date>
    <link>https://arxiv.org/abs/2012.11870</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[The problem of class imbalance is extensive for focusing on numerous applications in the real world. In such a situation, nearly all of the examples are labeled as one class called majority class, while far fewer examples are labeled as the other class usually, the more important class is called minority. Over the last few years, several types of research have been carried out on the issue of class imbalance, including data sampling, cost-sensitive analysis, Genetic Programming based models, bagging, boosting, etc. Nevertheless, in this survey paper, we enlisted the 24 related studies in the years 2003, 2008, 2010, 2012 and 2014 to 2019, focusing on the architecture of single, hybrid, and ensemble method design to understand the current status of improving classification output in machine learning techniques to fix problems with class imbalances. This survey paper also includes a statistical analysis of the classification algorithms under various methods and several other experimental conditions, as well as datasets used in different research papers.
]]></description>
<dc:subject>statistics data-balancing classification synthetic-data rather-interesting performance-measure to-write-about consider:sampling</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:32c0c43af08e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:data-balancing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:classification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:synthetic-data"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:sampling"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2005.02152">
    <title>[2005.02152] Augmented Semantic Signatures of Airborne LiDAR Point Clouds for Comparison</title>
    <dc:date>2020-09-23T14:27:19+00:00</dc:date>
    <link>https://arxiv.org/abs/2005.02152</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[LiDAR point clouds provide rich geometric information, which is particularly useful for the analysis of complex scenes of urban regions. Finding structural and semantic differences between two different three-dimensional point clouds, say, of the same region but acquired at different time instances is an important problem. A comparison of point clouds involves computationally expensive registration and segmentation. We are interested in capturing the relative differences in the geometric uncertainty and semantic content of the point cloud without the registration process. Hence, we propose an orientation-invariant geometric signature of the point cloud, which integrates its probabilistic geometric and semantic classifications. We study different properties of the geometric signature, which are an image-based encoding of geometric uncertainty and semantic content. We explore different metrics to determine differences between these signatures, which in turn compare point clouds without performing point-to-point registration. Our results show that the differences in the signatures corroborate with the geometric and semantic differences of the point clouds.
]]></description>
<dc:subject>computer-vision image-processing image-segmentation classification rather-interesting machine-learning performance-measure looking-to-see to-write-about consider:stochastic-resonance consider:noise-methids</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:e52b3e458b24/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computer-vision"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:image-processing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:image-segmentation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:classification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:stochastic-resonance"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:noise-methids"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1704.07067">
    <title>[1704.07067] Rerouting flows when links fail</title>
    <dc:date>2020-07-22T14:37:31+00:00</dc:date>
    <link>https://arxiv.org/abs/1704.07067</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We introduce and investigate reroutable flows, a robust version of network flows in which link failures can be mitigated by rerouting the affected flow. Given a capacitated network, a path flow is reroutable if after failure of an arbitrary arc, we can reroute the interrupted flow from the tail of that arc to the sink, without modifying the flow that is not affected by the failure. Similar types of restoration, which are often termed "local", were previously investigated in the context of network design, such as min-cost capacity planning. In this paper, our interest is in computing maximum flows under this robustness assumption. An important new feature of our model, distinguishing it from existing max robust flow models, is that no flow can get lost in the network. 
We also study a tightening of reroutable flows, called strictly reroutable flows, making more restrictive assumptions on the capacities available for rerouting. For both variants, we devise a reroutable-flow equivalent of an s-t-cut and show that the corresponding max flow/min cut gap is bounded by 2. It turns out that a strictly reroutable flow of maximum value can be found using a compact LP formulation, whereas the problem of finding a maximum reroutable flow is NP-hard, even when all capacities are in {1, 2}. However, the tightening can be used to get a 2-approximation for reroutable flows. This ratio is tight in general networks, but we show that in the case of unit capacities, every reroutable flow can be transformed into a strictly reroutable flow of same value. While it is NP-hard to compute a maximal integral flow even for unit capacities, we devise a surprisingly simple combinatorial algorithm that finds a half-integral strictly reroutable flow of value 1, or certifies that no such solutions exits. Finally, we also give a hardness result for the case of multiple arc failures.
]]></description>
<dc:subject>graph-theory algorithms network-theory operations-research optimization robustness performance-measure rather-interesting to-write-about to-simulate</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:51a64f47738d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:graph-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:network-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:operations-research"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:robustness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1902.07672">
    <title>[1902.07672] Non-asymptotic Analysis of Stochastic Methods for Non-Smooth Non-Convex Regularized Problems</title>
    <dc:date>2020-07-22T11:21:08+00:00</dc:date>
    <link>https://arxiv.org/abs/1902.07672</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Stochastic Proximal Gradient (SPG) methods have been widely used for solving optimization problems with a simple (possibly non-smooth) regularizer in machine learning and statistics. However, to the best of our knowledge no non-asymptotic convergence analysis of SPG exists for non-convex optimization with a non-smooth and non-convex regularizer. All existing non-asymptotic analysis of SPG for solving non-smooth non-convex problems require the non-smooth regularizer to be a convex function, and hence are not applicable to a non-smooth non-convex regularized problem. This work initiates the analysis to bridge this gap and opens the door to non-asymptotic convergence analysis of non-smooth non-convex regularized problems. We analyze several variants of mini-batch SPG methods for minimizing a non-convex objective that consists of a smooth non-convex loss and a non-smooth non-convex regularizer. Our contributions are two-fold: (i) we show that they enjoy the same complexities as their counterparts for solving convex regularized non-convex problems in terms of finding an approximate stationary point; (ii) we develop more practical variants using dynamic mini-batch size instead of a fixed mini-batch size without requiring the target accuracy level of solution. The significance of our results is that they improve upon the-state-of-art results for solving non-smooth non-convex regularized problems. We also empirically demonstrate the effectiveness of the considered SPG methods in comparison with other peer stochastic methods.
]]></description>
<dc:subject>performance-measure guessing rather-interesting statistics numerical-methods algorithms machine-learning horse-races to-understand to-simulate consider:lexicase consider:comparisons-in-GP</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:f5aabb982183/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:guessing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:horse-races"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:lexicase"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:comparisons-in-GP"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1610.06940">
    <title>[1610.06940] Safety Verification of Deep Neural Networks</title>
    <dc:date>2020-07-21T17:24:22+00:00</dc:date>
    <link>https://arxiv.org/abs/1610.06940</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Deep neural networks have achieved impressive experimental results in image classification, but can surprisingly be unstable with respect to adversarial perturbations, that is, minimal changes to the input image that cause the network to misclassify it. With potential applications including perception modules and end-to-end controllers for self-driving cars, this raises concerns about their safety. We develop a novel automated verification framework for feed-forward multi-layer neural networks based on Satisfiability Modulo Theory (SMT). We focus on safety of image classification decisions with respect to image manipulations, such as scratches or changes to camera angle or lighting conditions that would result in the same class being assigned by a human, and define safety for an individual decision in terms of invariance of the classification within a small neighbourhood of the original image. We enable exhaustive search of the region by employing discretisation, and propagate the analysis layer by layer. Our method works directly with the network code and, in contrast to existing methods, can guarantee that adversarial examples, if they exist, are found for the given region and family of manipulations. If found, adversarial examples can be shown to human testers and/or used to fine-tune the network. We implement the techniques using Z3 and evaluate them on state-of-the-art networks, including regularised and deep learning networks. We also compare against existing techniques to search for adversarial examples and estimate network robustness.
]]></description>
<dc:subject>robustness neural-networks validation rather-interesting performance-measure to-write-about consider:adjacency consider:performance-measures</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:f05266177bfe/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:robustness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:validation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:adjacency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:performance-measures"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1904.02063">
    <title>[1904.02063] Generalized Variational Inference: Three arguments for deriving new Posteriors</title>
    <dc:date>2020-05-22T21:16:22+00:00</dc:date>
    <link>https://arxiv.org/abs/1904.02063</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We advocate an optimization-centric view on and introduce a novel generalization of Bayesian inference. Our inspiration is the representation of Bayes' rule as infinite-dimensional optimization problem (Csiszar, 1975; Donsker and Varadhan; 1975, Zellner; 1988). First, we use it to prove an optimality result of standard Variational Inference (VI): Under the proposed view, the standard Evidence Lower Bound (ELBO) maximizing VI posterior is preferable to alternative approximations of the Bayesian posterior. Next, we argue for generalizing standard Bayesian inference. The need for this arises in situations of severe misalignment between reality and three assumptions underlying standard Bayesian inference: (1) Well-specified priors, (2) well-specified likelihoods, (3) the availability of infinite computing power. Our generalization addresses these shortcomings with three arguments and is called the Rule of Three (RoT). We derive it axiomatically and recover existing posteriors as special cases, including the Bayesian posterior and its approximation by standard VI. In contrast, approximations based on alternative ELBO-like objectives violate the axioms. Finally, we study a special case of the RoT that we call Generalized Variational Inference (GVI). GVI posteriors are a large and tractable family of belief distributions specified by three arguments: A loss, a divergence and a variational family. GVI posteriors have appealing properties, including consistency and an interpretation as approximate ELBO. The last part of the paper explores some attractive applications of GVI in popular machine learning models, including robustness and more appropriate marginals. After deriving black box inference schemes for GVI posteriors, their predictive performance is investigated on Bayesian Neural Networks and Deep Gaussian Processes, where GVI can comprehensively improve upon existing methods.
]]></description>
<dc:subject>models-and-modes to-understand probability-theory machine-learning define-your-terms performance-measure</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:ae6c80a8d812/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:models-and-modes"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:probability-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:define-your-terms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1905.03330">
    <title>[1905.03330] Universal Sound Separation</title>
    <dc:date>2020-05-21T11:49:53+00:00</dc:date>
    <link>https://arxiv.org/abs/1905.03330</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Recent deep learning approaches have achieved impressive performance on speech enhancement and separation tasks. However, these approaches have not been investigated for separating mixtures of arbitrary sounds of different types, a task we refer to as universal sound separation, and it is unknown how performance on speech tasks carries over to non-speech tasks. To study this question, we develop a dataset of mixtures containing arbitrary sounds, and use it to investigate the space of mask-based separation architectures, varying both the overall network architecture and the framewise analysis-synthesis basis for signal transformations. These network architectures include convolutional long short-term memory networks and time-dilated convolution stacks inspired by the recent success of time-domain enhancement networks like ConvTasNet. For the latter architecture, we also propose novel modifications that further improve separation performance. In terms of the framewise analysis-synthesis basis, we explore both a short-time Fourier transform (STFT) and a learnable basis, as used in ConvTasNet. For both of these bases, we also examine the effect of window size. In particular, for STFTs, we find that longer windows (25-50 ms) work best for speech/non-speech separation, while shorter windows (2.5 ms) work best for arbitrary sounds. For learnable bases, shorter windows (2.5 ms) work best on all tasks. Surprisingly, for universal sound separation, STFTs outperform learnable bases. Our best methods produce an improvement in scale-invariant signal-to-distortion ratio of over 13 dB for speech/non-speech separation and close to 10 dB for universal sound separation.
]]></description>
<dc:subject>signal-processing deep-learning neural-networks performance-measure benchmarks rather-interesting to-simulate to-write-about consider:training-cases consider:robustness</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:9934812ae053/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:signal-processing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:deep-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:benchmarks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:training-cases"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:robustness"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1803.00854">
    <title>[1803.00854] A more globally accurate dimensionality reduction method using triplets</title>
    <dc:date>2020-05-18T22:03:05+00:00</dc:date>
    <link>https://arxiv.org/abs/1803.00854</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We first show that the commonly used dimensionality reduction (DR) methods such as t-SNE and LargeVis poorly capture the global structure of the data in the low dimensional embedding. We show this via a number of tests for the DR methods that can be easily applied by any practitioner to the dataset at hand. Surprisingly enough, t-SNE performs the best w.r.t. the commonly used measures that reward the local neighborhood accuracy such as precision-recall while having the worst performance in our tests for global structure. We then contrast the performance of these two DR method against our new method called TriMap. The main idea behind TriMap is to capture higher orders of structure with triplet information (instead of pairwise information used by t-SNE and LargeVis), and to minimize a robust loss function for satisfying the chosen triplets. We provide compelling experimental evidence on large natural datasets for the clear advantage of the TriMap DR results. As LargeVis, TriMap scales linearly with the number of data points.
]]></description>
<dc:subject>dimension-reduction rather-interesting performance-measure representation to-write-about to-try consider:GP-population consider:lexicase</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:7d35de99ea5c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:dimension-reduction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:representation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-try"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:GP-population"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:lexicase"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1904.04412">
    <title>[1904.04412] 3D Quantum Cuts for Automatic Segmentation of Porous Media in Tomography Images</title>
    <dc:date>2020-05-18T21:52:44+00:00</dc:date>
    <link>https://arxiv.org/abs/1904.04412</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Binary segmentation of volumetric images of porous media is a crucial step towards gaining a deeper understanding of the factors governing biogeochemical processes at minute scales. Contemporary work primarily revolves around primitive techniques based on global or local adaptive thresholding that have known common drawbacks in image segmentation. Moreover, absence of a unified benchmark prohibits quantitative evaluation, which further clouds the impact of existing methodologies. In this study, we tackle the issue on both fronts. Firstly, by drawing parallels with natural image segmentation, we propose a novel, and automatic segmentation technique, 3D Quantum Cuts (QCuts-3D) grounded on a state-of-the-art spectral clustering technique. Secondly, we curate and present a publicly available dataset of 68 multiphase volumetric images of porous media with diverse solid geometries, along with voxel-wise ground truth annotations for each constituting phase. We provide comparative evaluations between QCuts-3D and the current state-of-the-art over this dataset across a variety of evaluation metrics. The proposed systematic approach achieves a 26% increase in AUROC while achieving a substantial reduction of the computational complexity of the state-of-the-art competitors. Moreover, statistical analysis reveals that the proposed method exhibits significant robustness against the compositional variations of porous media.
]]></description>
<dc:subject>tomography inverse-problems image-segmentation medical-technology algorithms image-analysis to-simulate to-write-about horse-races performance-measure compare:semantic-classification</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:55ba9624fe65/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:tomography"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:inverse-problems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:image-segmentation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:medical-technology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:image-analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:horse-races"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:compare:semantic-classification"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1805.01045">
    <title>[1805.01045] Alpha-Beta Divergence For Variational Inference</title>
    <dc:date>2020-05-06T11:32:44+00:00</dc:date>
    <link>https://arxiv.org/abs/1805.01045</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[This paper introduces a variational approximation framework using direct optimization of what is known as the {\it scale invariant Alpha-Beta divergence} (sAB divergence). This new objective encompasses most variational objectives that use the Kullback-Leibler, the R{é}nyi or the gamma divergences. It also gives access to objective functions never exploited before in the context of variational inference. This is achieved via two easy to interpret control parameters, which allow for a smooth interpolation over the divergence space while trading-off properties such as mass-covering of a target distribution and robustness to outliers in the data. Furthermore, the sAB variational objective can be optimized directly by repurposing existing methods for Monte Carlo computation of complex variational objectives, leading to estimates of the divergence instead of variational lower bounds. We show the advantages of this objective on Bayesian models for regression problems.
]]></description>
<dc:subject>machine-learning performance-measure multiobjective-optimization MO-but-what-if-we-collapse-them? to-write-about to-sigh-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:c6c3bf38e58f/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:multiobjective-optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:MO-but-what-if-we-collapse-them?"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-sigh-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.pnas.org/content/117/1/43">
    <title>Mechanics unlocks the morphogenetic puzzle of interlocking bivalved shells | PNAS</title>
    <dc:date>2020-05-02T14:51:20+00:00</dc:date>
    <link>https://www.pnas.org/content/117/1/43</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Brachiopods and mollusks are 2 shell-bearing phyla that diverged from a common shell-less ancestor more than 540 million years ago. Brachiopods and bivalve mollusks have also convergently evolved a bivalved shell that displays an apparently mundane, yet striking feature from a developmental point of view: When the shell is closed, the 2 valve edges meet each other in a commissure that forms a continuum with no gaps or overlaps despite the fact that each valve, secreted by 2 mantle lobes, may present antisymmetric ornamental patterns of varying regularity and size. Interlocking is maintained throughout the entirety of development, even when the shell edge exhibits significant irregularity due to injury or other environmental influences, which suggests a dynamic physical process of pattern formation that cannot be genetically specified. Here, we derive a mathematical framework, based on the physics of shell growth, to explain how this interlocking pattern is created and regulated by mechanical instabilities. By close consideration of the geometry and mechanics of 2 lobes of the mantle, constrained both by the rigid shell that they secrete and by each other, we uncover the mechanistic basis for the interlocking pattern. Our modeling framework recovers and explains a large diversity of shell forms and highlights how parametric variations in the growth process result in morphological variation. Beyond the basic interlocking mechanism, we also consider the intricate and striking multiscale-patterned edge in certain brachiopods. We show that this pattern can be explained as a secondary instability that matches morphological trends and data.

]]></description>
<dc:subject>developmental-biology evo-devo rather-interesting pattern-formation theoretical-biology dynamical-systems performance-measure to-write-about to-simulate</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:4fb2b291258e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:developmental-biology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:evo-devo"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:pattern-formation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:theoretical-biology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:dynamical-systems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2002.08512">
    <title>[2002.08512] The Problem with Metrics is a Fundamental Problem for AI</title>
    <dc:date>2020-05-02T14:46:31+00:00</dc:date>
    <link>https://arxiv.org/abs/2002.08512</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Optimizing a given metric is a central aspect of most current AI approaches, yet overemphasizing metrics leads to manipulation, gaming, a myopic focus on short-term goals, and other unexpected negative consequences. This poses a fundamental contradiction for AI development. Through a series of real-world case studies, we look at various aspects of where metrics go wrong in practice and aspects of how our online environment and current business practices are exacerbating these failures. Finally, we propose a framework towards mitigating the harms caused by overemphasis of metrics within AI by: (1) using a slate of metrics to get a fuller and more nuanced picture, (2) combining metrics with qualitative accounts, and (3) involving a range of stakeholders, including those who will be most impacted.
]]></description>
<dc:subject>philosophy-of-engineering machine-learning performance-measure what-gets-measured-gets-fudged to-write-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:69f1dff9c10d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:philosophy-of-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:what-gets-measured-gets-fudged"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://en.wikipedia.org/wiki/RANDU">
    <title>RANDU - Wikipedia</title>
    <dc:date>2020-05-02T12:31:54+00:00</dc:date>
    <link>https://en.wikipedia.org/wiki/RANDU</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In general, when an LCG with modulus 231 is used to produce points (xk, xk+1, xk+2) in 3-dimensional space, the points fall into no more than 2,344 parallel planes[5], a result which indicates an LCG is unsuitable for Monte Carlo simulation. Choice of multiplier determines the number of planes. To show the problem with the values of multiplier 65539 and modulus 231 chosen for RANDU, consider the following calculation where every term should be taken mod 231. Start by writing the recursive relation as:
]]></description>
<dc:subject>randomness algorithms pseudorandom-number-generator numerical-methods performance-measure bad-design to-write-about to-simulate consider:genetic-programming</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:5c61b299eaf5/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:randomness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:pseudorandom-number-generator"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:bad-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:genetic-programming"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1812.07270">
    <title>[1812.07270] AVATAR : Fixing Semantic Bugs with Fix Patterns of Static Analysis Violations</title>
    <dc:date>2020-03-20T16:38:42+00:00</dc:date>
    <link>https://arxiv.org/abs/1812.07270</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Fix pattern-based patch generation is a promising direction in Automated Program Repair (APR). Notably, it has been demonstrated to produce more acceptable and correct patches than the patches obtained with mutation operators through genetic programming. The performance of pattern-based APR systems, however, depends on the fix ingredients mined from fix changes in development histories. Unfortunately, collecting a reliable set of bug fixes in repositories can be challenging. In this paper, we propose to investigate the possibility in an APR scenario of leveraging code changes that address violations by static bug detection tools. To that end, we build the AVATAR APR system, which exploits fix patterns of static analysis violations as ingredients for patch generation. Evaluated on the Defects4J benchmark, we show that, assuming a perfect localization of faults, AVATAR can generate correct patches to fix 34/39 bugs. We further find that AVATAR yields performance metrics that are comparable to that of the closely-related approaches in the literature. While AVATAR outperforms many of the state-of-the-art pattern-based APR systems, it is mostly complementary to current approaches. Overall, our study highlights the relevance of static bug finding tools as indirect contributors of fix ingredients for addressing code defects identified with functional test cases.
]]></description>
<dc:subject>refactoring software-engineering performance-measure rather-interesting rewriting-systems to-download to-simulate consider:refactoring-as-such</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:be090c45941a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:refactoring"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:software-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rewriting-systems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-download"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:refactoring-as-such"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.56.5609">
    <title>CiteSeerX — RPG - Heuristics for the Generation of Random Polygons</title>
    <dc:date>2020-01-30T16:52:43+00:00</dc:date>
    <link>http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.56.5609</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We consider the problem of randomly generating simple and star-shaped polygons on a given set of points. This problem is of considerable importance in the practical evaluation of algorithms that operate on polygons, where it is necessary to check the correctness and to determine the actual CPU-consumption of an algorithm experimentally. Since no polynomial-time solution for the uniformly random generation of polygons is known, we present and analyze several heuristics. All heuristics described in this paper have been implemented and are part of our RandomPolygonGenerator, RPG. We have tested all heuristics, and report experimental results on their CPU-consumption, their quality, and their characteristics. RPG is publically available via http://www.cosy.sbg.ac.at/~held/projects/rpg/rpg.html. 1 Introduction In this paper 1 we deal with the random generation of simple polygons on a given set of points: Ideally, given a set S = fs 1 ; : : : ; s n g of n points, we would like to generat...
]]></description>
<dc:subject>probability-theory computational-geometry sampling rather-interesting performance-measure to-simulate to-write-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:3adbf07e0b0c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:probability-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computational-geometry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:sampling"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1812.05163">
    <title>[1812.05163] Declination as a Metric to Detect Partisan Gerrymandering</title>
    <dc:date>2020-01-27T23:32:12+00:00</dc:date>
    <link>https://arxiv.org/abs/1812.05163</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We explore the Declination, a new metric intended to detect partisan gerrymandering. We consider instances in which each district has equal turnout, the maximum turnout to minimum turnout is bounded, and turnout is unrestricted. For each of these cases, we show exactly which vote-share, seat-share pairs (V,S) have an election outcome with Declination equal to 0. We also show how our analyses can be applied to finding vote-share, seat-share pairs that are possible for nonzero Declination. 
Within our analyses, we show that Declination cannot detect all forms of packing and cracking, and we compare the Declination to the Efficiency Gap. We show that these two metrics can behave quite differently, and give explicit examples of that occurring.
]]></description>
<dc:subject>statistics politics gerrymandering cultural-engineering rather-interesting performance-measure to-write-about to-simulate</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:76021792de5c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:politics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:gerrymandering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:cultural-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1702.08066">
    <title>[1702.08066] On the Classification and Algorithmic Analysis of Carmichael Numbers</title>
    <dc:date>2020-01-23T11:15:16+00:00</dc:date>
    <link>https://arxiv.org/abs/1702.08066</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In this paper, we study the properties of Carmichael numbers, false positives to several primality tests. We provide a classification for Carmichael numbers with a proportion of Fermat witnesses of less than 50%, based on if the smallest prime factor is greater than a determined lower bound. In addition, we conduct a Monte Carlo simulation as part of a probabilistic algorithm to detect if a given composite number is Carmichael. We modify this highly accurate algorithm with a deterministic primality test to create a novel, more efficient algorithm that differentiates between Carmichael numbers and prime numbers.
]]></description>
<dc:subject>number-theory primes rather-interesting feature-construction classification tricky-cases edge-cases algorithms performance-measure to-simulate to-write-about consider:classification computational-complexity</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:162b3033bb49/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:number-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:primes"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:feature-construction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:classification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:tricky-cases"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:edge-cases"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:classification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computational-complexity"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1909.10140">
    <title>[1909.10140] A new coefficient of correlation</title>
    <dc:date>2019-11-23T12:50:10+00:00</dc:date>
    <link>https://arxiv.org/abs/1909.10140</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Is it possible to define a coefficient of correlation which is (a) as simple as the classical coefficients like Pearson's correlation or Spearman's correlation, and yet (b) consistently estimates some simple and interpretable measure of the degree of dependence between the variables, which is 0 if and only if the variables are independent and 1 if and only if one is a measurable function of the other, and (c) has a simple asymptotic theory under the hypothesis of independence, like the classical coefficients? This article answers this question in the affirmative, by producing such a coefficient. No assumptions are needed on the distributions of the variables. There are several coefficients in the literature that converge to 0 if and only if the variables are independent, but none that satisfy any of the other properties mentioned above.
]]></description>
<dc:subject>statistics probability-theory performance-measure rather-interesting nudge-targets to-write-about to-replicate consider:performance-measures</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:9623577ab990/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:probability-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:nudge-targets"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-replicate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:performance-measures"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1802.00296">
    <title>[1802.00296] $S$-Leaping: An adaptive, accelerated stochastic simulation algorithm, bridging $τ$-leaping and $R$-leaping</title>
    <dc:date>2019-11-03T11:48:18+00:00</dc:date>
    <link>https://arxiv.org/abs/1802.00296</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We propose the S-leaping algorithm for the acceleration of Gillespie's stochastic simulation algorithm that combines the advantages of the two main accelerated methods; the τ-leaping and R-leaping algorithms. These algorithms are known to be efficient under different conditions; the τ-leaping is efficient for non-stiff systems or systems with partial equilibrium, while the R-leaping performs better in stiff system thanks to an efficient sampling procedure. However, even a small change in a system's set up can critically affect the nature of the simulated system and thus reduce the efficiency of an accelerated algorithm. The proposed algorithm combines the efficient time step selection from the τ-leaping with the effective sampling procedure from the R-leaping algorithm. The S-leaping is shown to maintain its efficiency under different conditions and in the case of large and stiff systems or systems with fast dynamics, the S-leaping outperforms both methods. We demonstrate the performance and the accuracy of the S-leaping in comparison with the τ-leaping and R-leaping on a number of benchmark systems involving biological reaction networks.
]]></description>
<dc:subject>simulation numerical-methods Markov-models Monte-Carlo-models probability-theory algorithms horse-races rather-interesting to-write-about performance-measure approximation</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:c4efe05241f4/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:simulation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:Markov-models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:Monte-Carlo-models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:probability-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:horse-races"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:approximation"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1602.04181">
    <title>[1602.04181] Spectral Alignment of Graphs</title>
    <dc:date>2019-09-28T10:51:59+00:00</dc:date>
    <link>https://arxiv.org/abs/1602.04181</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Graph alignment refers to the problem of finding a bijective mapping across vertices of two graphs such that, if two nodes are connected in the first graph, their images are connected in the second graph. This problem arises in many fields such as computational biology, social sciences, and computer vision and is often cast as a quadratic assignment problem (QAP). Most standard graph alignment methods consider an optimization that maximizes the number of matches between the two graphs, ignoring the effect of mismatches. We propose a generalized graph alignment formulation that considers both matches and mismatches in a standard QAP formulation. This modification can have a major impact in aligning graphs with different sizes and heterogenous edge densities. Moreover, we propose two methods for solving the generalized graph alignment problem based on spectral decomposition of matrices. We compare the performance of proposed methods with some existing graph alignment algorithms including Natalie2, GHOST, IsoRank, NetAlign, Klau's approach as well as a semidefinite programming-based method over various synthetic and real graph models. Our proposed method based on simultaneous alignment of multiple eigenvectors leads to consistently good performance in different graph models. In particular, in the alignment of regular graph structures which is one of the most difficult graph alignment cases, our proposed method significantly outperforms other methods.
]]></description>
<dc:subject>graph-theory similarity-measures clustering distance numerical-methods algorithms performance-measure rather-interesting consider:looking-to-see consider:robustness to-simulate</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:d47a66c1988f/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:graph-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:similarity-measures"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:clustering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:distance"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:robustness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1902.00415">
    <title>[1902.00415] Normalized Wasserstein Distance for Mixture Distributions with Applications in Adversarial Learning and Domain Adaptation</title>
    <dc:date>2019-09-28T10:47:41+00:00</dc:date>
    <link>https://arxiv.org/abs/1902.00415</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Understanding proper distance measures between distributions is at the core of several learning tasks such as generative models, domain adaptation, clustering, etc. In this work, we focus on {\it mixture distributions} that arise naturally in several application domains where the data contains different sub-populations. For mixture distributions, established distance measures such as the Wasserstein distance do not take into account imbalanced mixture proportions. Thus, even if two mixture distributions have identical mixture components but different mixture proportions, the Wasserstein distance between them will be large. This often leads to undesired results in distance-based learning methods for mixture distributions. In this paper, we resolve this issue by introducing {\it Normalized Wasserstein} distance. The key idea is to introduce mixture proportions as optimization variables, effectively normalizing mixture proportions in the Wasserstein formulation. Using the proposed normalized Wasserstein distance, instead of the vanilla one, leads to significant gains working with mixture distributions with imbalanced mixture proportions. We demonstrate effectiveness of the proposed distance in GANs, domain adaptation, adversarial clustering and hypothesis testing over mixture of Gaussians, MNIST, CIFAR-10, CelebA and VISDA datasets.
]]></description>
<dc:subject>statistics distance metrics performance-measure numerical-methods dimension-reduction to-understand machine-learning to-simulate</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:5cc0f0836a46/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:distance"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metrics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:dimension-reduction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1909.10140#">
    <title>[1909.10140] A new coefficient of correlation</title>
    <dc:date>2019-09-28T10:45:13+00:00</dc:date>
    <link>https://arxiv.org/abs/1909.10140#</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Is it possible to define a coefficient of correlation which is (a) as simple as the classical coefficients like Pearson's correlation or Spearman's correlation, and yet (b) consistently estimates some simple and interpretable measure of the degree of dependence between the variables, which is 0 if and only if the variables are independent and 1 if and only if one is a measurable function of the other, and (c) has a simple asymptotic theory under the hypothesis of independence, like the classical coefficients? This article answers this question in the affirmative, by producing such a coefficient. No assumptions are needed on the distributions of the variables. There are several coefficients in the literature that converge to 0 if and only if the variables are independent, but none that satisfy any of the other properties mentioned above.
]]></description>
<dc:subject>statistics algorithms rather-interesting engineering-design performance-measure define-your-terms to-simulate to-write-about consider:rediscovery</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:56410294020d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:define-your-terms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:rediscovery"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1903.09139">
    <title>[1903.09139] Harmless interpolation of noisy data in regression</title>
    <dc:date>2019-09-22T12:03:26+00:00</dc:date>
    <link>https://arxiv.org/abs/1903.09139</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[A continuing mystery in understanding the empirical success of deep neural networks is their ability to achieve zero training error and generalize well, even when the training data is noisy and there are more parameters than data points. We investigate this overparameterized regime in linear regression, where all solutions that minimize training error interpolate the data, including noise. We characterize the fundamental generalization (mean-squared) error of any interpolating solution in the presence of noise, and show that this error decays to zero with the number of features. Thus, overparameterization can be explicitly beneficial in ensuring harmless interpolation of noise. We discuss two root causes for poor generalization that are complementary in nature -- signal "bleeding" into a large number of alias features, and overfitting of noise by parsimonious feature selectors. For the sparse linear model with noise, we provide a hybrid interpolating scheme that mitigates both these issues and achieves order-optimal MSE over all possible interpolating solutions.
]]></description>
<dc:subject>statistics models performance-measure interpolation rather-interesting to-write-about consider:symbolic-regression consider:aggregate-model-performance consider:lexicase</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:a33df4020411/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:interpolation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:symbolic-regression"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:aggregate-model-performance"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:lexicase"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/math/0402212">
    <title>[math/0402212] Criticality for the Gehring link problem</title>
    <dc:date>2019-09-11T11:51:30+00:00</dc:date>
    <link>https://arxiv.org/abs/math/0402212</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In 1974, Gehring posed the problem of minimizing the length of two linked curves separated by unit distance. This constraint can be viewed as a measure of thickness for links, and the ratio of length over thickness as the ropelength. In this paper we refine Gehring's problem to deal with links in a fixed link-homotopy class: we prove ropelength minimizers exist and introduce a theory of ropelength criticality. 
Our balance criterion is a set of necessary and sufficient conditions for criticality, based on a strengthened, infinite-dimensional version of the Kuhn--Tucker theorem. We use this to prove that every critical link is C^1 with finite total curvature. The balance criterion also allows us to explicitly describe critical configurations (and presumed minimizers) for many links including the Borromean rings. We also exhibit a surprising critical configuration for two clasped ropes: near their tips the curvature is unbounded and a small gap appears between the two components. These examples reveal the depth and richness hidden in Gehring's problem and our natural extension.
]]></description>
<dc:subject>rather-interesting knot-theory optimization numerical-methods engineering-design performance-measure to-write-about to-simulate consider:constraint-satisfaction consider:feature-discovery</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:1f431e5bdb57/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:knot-theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:numerical-methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:constraint-satisfaction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:feature-discovery"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://dl.acm.org/citation.cfm?id=2601224">
    <title>The Connect-The-Dots family of puzzles</title>
    <dc:date>2019-09-10T13:05:40+00:00</dc:date>
    <link>https://dl.acm.org/citation.cfm?id=2601224</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[In this paper we introduce several innovative variants on the classic Connect-The-Dots puzzle. We study the underlying geometric principles and investigate methods for the automatic generation of high-quality puzzles from line drawings.
Specifically, we introduce three new variants of the classic Connect-The-Dots puzzle. These new variants use different rules for drawing connections, and have several advantages: no need for printed numbers in the puzzle (which look ugly in the final drawing), and perhaps more challenging "game play", making the puzzles suitable for different age groups. We study the rules of all four variants in the family, and design principles describing what makes a good puzzle. We identify general principles that apply across the different variants, as well as specific implementations of those principles in the different variants. We make these mathematically precise in the form of criteria a puzzle should satisfy.
Furthermore, we investigate methods for the automatic generation of puzzles from a plane graph that describes the input drawing. We show that the problem of generating a good puzzle --one satisfying the mentioned criteria-- is computationally hard, and present several heuristic algorithms.
Using our implementation for generating puzzles, we evaluate the quality of the resulting puzzles with respect to two parameters: one for similarity to the original line drawing, and one for ambiguity; i.e. what is the visual accuracy needed to solve the puzzle.
]]></description>
<dc:subject>image-processing feature-construction algorithms rather-interesting optimization performance-measure consider:looking-to-see consider:representation</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:11bdf7491de7/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:image-processing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:feature-construction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:representation"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1708.01559">
    <title>[1708.01559] Spherical Geometry and the Least Symmetric Triangle</title>
    <dc:date>2019-09-09T10:43:14+00:00</dc:date>
    <link>https://arxiv.org/abs/1708.01559</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We study the problem of determining the least symmetric triangle, which arises both from pure geometry and from the study of molecular chirality in chemistry. Using the correspondence between planar n-gons and points in the Grassmannian of 2-planes in real n-space introduced by Hausmann and Knutson, this corresponds to finding the point in the fundamental domain of the hyperoctahedral group action on the Grassmannian which is furthest from the boundary, which we compute exactly. We also determine the least symmetric obtuse and acute triangles. These calculations provide prototypes for computations on polygon and shape spaces.
]]></description>
<dc:subject>symmetry optimization rather-interesting define-your-terms performance-measure to-write-about to-emulate to-simulate consider:looking-to-see consider:robustness consider:stamp-collecting geometry plane-geometry</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:57d2fd877722/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:symmetry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:define-your-terms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-emulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:robustness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:stamp-collecting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:geometry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:plane-geometry"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1803.02015">
    <title>[1803.02015] Generative Modeling of Multimodal Multi-Human Behavior</title>
    <dc:date>2019-09-08T14:50:29+00:00</dc:date>
    <link>https://arxiv.org/abs/1803.02015</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[This work presents a methodology for modeling and predicting human behavior in settings with N humans interacting in highly multimodal scenarios (i.e. where there are many possible highly-distinct futures). A motivating example includes robots interacting with humans in crowded environments, such as self-driving cars operating alongside human-driven vehicles or human-robot collaborative bin packing in a warehouse. Our approach to model human behavior in such uncertain environments is to model humans in the scene as nodes in a graphical model, with edges encoding relationships between them. For each human, we learn a multimodal probability distribution over future actions from a dataset of multi-human interactions. Learning such distributions is made possible by recent advances in the theory of conditional variational autoencoders and deep learning approximations of probabilistic graphical models. Specifically, we learn action distributions conditioned on interaction history, neighboring human behavior, and candidate future agent behavior in order to take into account response dynamics. We demonstrate the performance of such a modeling approach in modeling basketball player trajectories, a highly multimodal, multi-human scenario which serves as a proxy for many robotic applications.
]]></description>
<dc:subject>swarms generative-models robotics to-understand performance-measure consider:performance-measures consider:genetic-programming to-simulate collective-behavior planning</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:6a22d37a283a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:swarms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:generative-models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:robotics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:performance-measures"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:genetic-programming"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-simulate"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:collective-behavior"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:planning"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1509.02241">
    <title>[1509.02241] The local optimality of the double lattice packing</title>
    <dc:date>2019-09-07T12:06:46+00:00</dc:date>
    <link>https://arxiv.org/abs/1509.02241</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[This paper introduces a technique for proving the local optimality of packing configurations. Applying this technique to a general convex polygon, we prove that the construction of the optimal double lattice packing by Kuperberg and Kuperberg is also locally optimal in the full space of packings.
]]></description>
<dc:subject>geometry proof techniques algorithms packing plane-geometry optimization performance-measure to-understand to-write-about consider:looking-to-see consider:rediscovery consider:robustness</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:7caa3f823a4d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:geometry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:proof"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:techniques"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:packing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:plane-geometry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:rediscovery"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:robustness"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1608.07617">
    <title>[1608.07617] &quot;Sampling&quot;' as a Baseline Optimizer for Search-based Software Engineering</title>
    <dc:date>2019-08-30T10:59:42+00:00</dc:date>
    <link>https://arxiv.org/abs/1608.07617</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Increasingly, Software Engineering (SE) researchers use search-based optimization techniques to solve SE problems with multiple conflicting objectives. These techniques often apply CPU-intensive evolutionary algorithms to explore generations of mutations to a population of candidate solutions. An alternative approach, proposed in this paper, is to start with a very large population and sample down to just the better solutions. We call this method "SWAY", short for "the sampling way". Sway is very simple to implement and, in studies with various software engineering models, this sampling approach was found to be competitive with corresponding state-of-the-art evolutionary algorithms while requiring far less computation cost. Considering the simplicity and effectiveness of Sway, we, therefore, propose this approach as a baseline method for search-based software engineering models, especially for models that are very slow to execute.
]]></description>
<dc:subject>metaheuristics genetic-programming multiobjective-optimization evolutionary-algorithms performance-measure rather-interesting search-operators to-write-about to-replicate</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:d1ada658a31a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metaheuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:genetic-programming"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:multiobjective-optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:evolutionary-algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:search-operators"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-replicate"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1808.09325">
    <title>[1808.09325] &quot;Life never matters in the DEMOCRATS MIND&quot;: Examining Strategies of Retweeted Social Bots During a Mass Shooting Event</title>
    <dc:date>2019-08-30T10:55:54+00:00</dc:date>
    <link>https://arxiv.org/abs/1808.09325</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[This exploratory study examines the strategies of social bots on Twitter that were retweeted following a mass shooting event. Using a case study method to frame our work, we collected over seven million tweets during a one-month period following a mass shooting in Parkland, Florida. From this dataset, we selected retweets of content generated by over 400 social bot accounts to determine what strategies these bots were using and the effectiveness of these strategies as indicated by the number of retweets. We employed qualitative and quantitative methods to capture both macro- and micro-level perspectives. Our findings suggest that bots engage in more diverse strategies than solely waging disinformation campaigns, including baiting and sharing information. Further, we found that while bots amplify conversation about mass shootings, humans were primarily responsible for disseminating bot-generated content. These findings add depth to the current understanding of bot strategies and their effectiveness. Understanding these strategies can inform efforts to combat dubious information as well as more insidious disinformation campaigns.
]]></description>
<dc:subject>social-media social-engineering botnets propaganda rather-interesting looking-to-see performance-measure consider:countermeasures consider:chaffnet</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:055f0840c70f/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:social-media"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:social-engineering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:botnets"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:propaganda"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:countermeasures"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:consider:chaffnet"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1711.08042">
    <title>[1711.08042] &quot;I know it when I see it&quot;. Visualization and Intuitive Interpretability</title>
    <dc:date>2019-08-30T10:53:34+00:00</dc:date>
    <link>https://arxiv.org/abs/1711.08042</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Most research on the interpretability of machine learning systems focuses on the development of a more rigorous notion of interpretability. I suggest that a better understanding of the deficiencies of the intuitive notion of interpretability is needed as well. I show that visualization enables but also impedes intuitive interpretability, as it presupposes two levels of technical pre-interpretation: dimensionality reduction and regularization. Furthermore, I argue that the use of positive concepts to emulate the distributed semantic structure of machine learning models introduces a significant human bias into the model. As a consequence, I suggest that, if intuitive interpretability is needed, singular representations of internal model states should be avoided.
]]></description>
<dc:subject>interpretability machine-learning philosophy-of-science define-your-terms performance-measure trade-offs to-write-about rather-interesting</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:70cd8ddc95f8/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:interpretability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:philosophy-of-science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:define-your-terms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:trade-offs"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1806.04236">
    <title>[1806.04236] &quot;AIded with emotions&quot; - a new design approach towards affective computer systems</title>
    <dc:date>2019-08-30T10:45:19+00:00</dc:date>
    <link>https://arxiv.org/abs/1806.04236</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[As technologies become more and more pervasive, there is a need for considering the affective dimension of interaction with computer systems to make them more human-like. Current demands for this matter include accurate emotion recognition, reliable emotion modeling, and use of unobtrusive, easily accessible and preferably wearable measurement devices. While AI methods provide many possibilities for better affective information processing, it is not a common scenario for both emotion recognition and modeling to be integrated in the design phase. To address this concern, we propose a new approach based on affective design patterns in the context of video games, together with summary of experiments conducted to test the preliminary hypotheses.
]]></description>
<dc:subject>affective-computing games engineering-design emotions performance-measure rather-interesting to-understand</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:e3cf7d5cff67/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:affective-computing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:games"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:emotions"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1902.11214">
    <title>[1902.11214] A Multilayer Structure Facilitates the Production of Antifragile Systems in Boolean Network Models</title>
    <dc:date>2019-08-06T13:54:14+00:00</dc:date>
    <link>https://arxiv.org/abs/1902.11214</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Antifragility is a property to not only resist stress and but also to benefit from it. Even though antifragile dynamics are found in various real-world complex systems where multiple subsystems interact with each other, the attribute has not been quantitatively explored yet in those complex systems which can be regarded as multilayer networks. Here we study how the multilayer structure affects the antifragility of the whole system. By comparing single-layer and multilayer Boolean networks based on our recently proposed antifragility measure, we found that the multilayer structure facilitated the production of antifragile systems. Our measure and findings can be utilized for many applications from understanding properties of biological systems with multilayer structures to designing more antifragile engineered systems.
]]></description>
<dc:subject>Talebism Kauffmania boolean-networks automata engineering-design emergent-design to-write-about to-clean-up performance-measure meh an-abstraction-doing-a-bit-too-much-work</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:9d47bec73e73/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:Talebism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:Kauffmania"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:boolean-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:automata"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:emergent-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-clean-up"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:meh"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:an-abstraction-doing-a-bit-too-much-work"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1107.5667">
    <title>[1107.5667] Fractal bodies invisible in 2 and 3 directions</title>
    <dc:date>2019-08-06T09:33:09+00:00</dc:date>
    <link>https://arxiv.org/abs/1107.5667</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[We study the problem of invisibility for bodies with a mirror surface in the framework of geometrical optics. We show that for any two given directions it is possible to construct a two-dimensional fractal body invisible in these directions. Moreover, there exists a three-dimensional fractal body invisible in three orthogonal directions. The work continues the previous study in [A. Aleksenko and A. Plakhov. Bodies of zero resistance and bodies invisible in one direction. Nonlinearity 22, 1247-1258 (2009)], [A Plakhov and V Roshchina. Invisibility in billiards. Nonlinearity 24, 847-854 (2011)], where two-dimensional bodies invisible in one direction and three-dimensional bodies invisible in one and two orthogonal directions were constructed.
]]></description>
<dc:subject>constraint-satisfaction billiards rather-interesting engineering-design open-problems performance-measure looking-to-see to-write-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:9a9259cb727d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:constraint-satisfaction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:billiards"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:open-problems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:looking-to-see"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1807.01672">
    <title>[1807.01672] Ranked Reward: Enabling Self-Play Reinforcement Learning for Combinatorial Optimization</title>
    <dc:date>2019-08-06T09:11:52+00:00</dc:date>
    <link>https://arxiv.org/abs/1807.01672</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Adversarial self-play in two-player games has delivered impressive results when used with reinforcement learning algorithms that combine deep neural networks and tree search. Algorithms like AlphaZero and Expert Iteration learn tabula-rasa, producing highly informative training data on the fly. However, the self-play training strategy is not directly applicable to single-player games. Recently, several practically important combinatorial optimisation problems, such as the travelling salesman problem and the bin packing problem, have been reformulated as reinforcement learning problems, increasing the importance of enabling the benefits of self-play beyond two-player games. We present the Ranked Reward (R2) algorithm which accomplishes this by ranking the rewards obtained by a single agent over multiple games to create a relative performance metric. Results from applying the R2 algorithm to instances of a two-dimensional and three-dimensional bin packing problems show that it outperforms generic Monte Carlo tree search, heuristic algorithms and integer programming solvers. We also present an analysis of the ranked reward mechanism, in particular, the effects of problem instances with varying difficulty and different ranking thresholds.
]]></description>
<dc:subject>machine-learning reinforcement-learning algorithms self-play mechanism-design optimization performance-measure to-understand operations-research metaheuristics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:062f6ce17047/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:reinforcement-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:self-play"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:mechanism-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:operations-research"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:metaheuristics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://ddd.fit.cvut.cz/prj/Benchmarks/index.php?page=download">
    <title>Collection of Digital Design Benchmarks</title>
    <dc:date>2019-07-29T11:03:27+00:00</dc:date>
    <link>https://ddd.fit.cvut.cz/prj/Benchmarks/index.php?page=download</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Collection of Digital Design Benchmarks]]></description>
<dc:subject>benchmarking engineering-design circuits genetic-programming constraint-satisfaction performance-measure to-write-about boolean-networks</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:db5180be6a39/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:benchmarking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:engineering-design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:circuits"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:genetic-programming"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:constraint-satisfaction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:boolean-networks"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1811.01910">
    <title>[1811.01910] Evolutionary Data Measures: Understanding the Difficulty of Text Classification Tasks</title>
    <dc:date>2019-07-25T11:03:05+00:00</dc:date>
    <link>https://arxiv.org/abs/1811.01910</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Classification tasks are usually analysed and improved through new model architectures or hyperparameter optimisation but the underlying properties of datasets are discovered on an ad-hoc basis as errors occur. However, understanding the properties of the data is crucial in perfecting models. In this paper we analyse exactly which characteristics of a dataset best determine how difficult that dataset is for the task of text classification. We then propose an intuitive measure of difficulty for text classification datasets which is simple and fast to calculate. We show that this measure generalises to unseen data by comparing it to state-of-the-art datasets and results. This measure can be used to analyse the precise source of errors in a dataset and allows fast estimation of how difficult a dataset is to learn. We searched for this measure by training 12 classical and neural network based models on 78 real-world datasets, then use a genetic algorithm to discover the best measure of difficulty. Our difficulty-calculating code ( this https URL ) and datasets ( this http URL ) are publicly available.
]]></description>
<dc:subject>machine-learning performance-measure performance-space-analysis clustering adversarial-testing to-write-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:db970d1f461e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-space-analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:clustering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:adversarial-testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1906.07774">
    <title>[1906.07774] Information matrices and generalization</title>
    <dc:date>2019-07-24T11:27:44+00:00</dc:date>
    <link>https://arxiv.org/abs/1906.07774</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[This work revisits the use of information criteria to characterize the generalization of deep learning models. In particular, we empirically demonstrate the effectiveness of the Takeuchi information criterion (TIC), an extension of the Akaike information criterion (AIC) for misspecified models, in estimating the generalization gap, shedding light on why quantities such as the number of parameters cannot quantify generalization. The TIC depends on both the Hessian of the loss H and the covariance of the gradients C. By exploring the similarities and differences between these two matrices as well as the Fisher information matrix F, we study the interplay between noise and curvature in deep models. We also address the question of whether C is a reasonable approximation to F, as is commonly assumed.]]></description>
<dc:subject>machine-learning neural-networks generalization performance-measure rather-interesting to-understand to-generalize (hah) to-write-about</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:b198aff18217/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:neural-networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:generalization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:rather-interesting"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-understand"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-generalize"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:(hah)"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-write-about"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1802.07481">
    <title>[1802.07481] Celer: a Fast Solver for the Lasso with Dual Extrapolation</title>
    <dc:date>2019-07-24T10:51:11+00:00</dc:date>
    <link>https://arxiv.org/abs/1802.07481</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[Convex sparsity-inducing regularizations are ubiquitous in high-dimensional machine learning, but solving the resulting optimization problems can be slow. To accelerate solvers, state-of-the-art approaches consist in reducing the size of the optimization problem at hand. In the context of regression, this can be achieved either by discarding irrelevant features (screening techniques) or by prioritizing features likely to be included in the support of the solution (working set techniques). Duality comes into play at several steps in these techniques. Here, we propose an extrapolation technique starting from a sequence of iterates in the dual that leads to the construction of improved dual points. This enables a tighter control of optimality as used in stopping criterion, as well as better screening performance of Gap Safe rules. Finally, we propose a working set strategy based on an aggressive use of Gap Safe screening rules. Thanks to our new dual point construction, we show significant computational speedups on multiple real-world problems.
]]></description>
<dc:subject>machine-learning statistics horse-races performance-measure computational-complexity to-read algorithms</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:ba797f2dc4e8/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:horse-races"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:computational-complexity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:to-read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:algorithms"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://hackingsemantics.xyz/2019/leaderboards/">
    <title>How the Transformers broke NLP leaderboards - Hacking semantics</title>
    <dc:date>2019-07-24T10:44:04+00:00</dc:date>
    <link>https://hackingsemantics.xyz/2019/leaderboards/</link>
    <dc:creator>Vaguery</dc:creator><description><![CDATA[If leaderboards are to highlight the actual progress, we need to incentivize new architectures rather than teams outspending each other. Obviously, huge pretrained models are valuable, but unless the authors show that their system consistently behaves differently from its competition with comparable data & compute, it is not clear whether they are presenting a model or a resource.

Furthermore, much of this research is not reproducible: nobody is going to spend $250,000 just to repeat XLNet training. Given the fact that its ablation study showed only 1-2% gain over BERT in 3 datasets out of 4 (Yang et al., 2019), we don’t actually know for sure that its masking strategy is more successful than BERT’s.

At the same time, the development of leaner models is dis-incentivized, as their task is fundamentally harder and the leaderboard-oriented community only rewards the SOTA. That, in its turn, prices out of competitions academic teams, which will not result in students becoming better engineers when they graduate.

Last but not the least, huge DL models are often overparametrized (Frankle & Carbin, 2019; Wu, Fan, Baevski, Dauphin, & Auli, 2019). As an example, the smaller version of BERT achieves better scores on a number of syntax-testing experiments than the larger one (Goldberg, 2019). The fact that DL models require a lot of compute is not necessarily a bad thing in itself, but wasting compute is not ideal for the environment (Strubell, Ganesh, & McCallum, 2019).

]]></description>
<dc:subject>benchmarking machine-learning horse-races performance-measure multiobjective-optimization (use-it)</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:Vaguery/b:bd8204972f3a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:benchmarking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:machine-learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:horse-races"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:performance-measure"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:multiobjective-optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:Vaguery/t:(use-it)"/>
</rdf:Bag></taxo:topics>
</item>
</rdf:RDF>