<?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 (cshalizi)</title>
    <link>https://pinboard.in/u:cshalizi/public/</link>
    <description>recent bookmarks from cshalizi</description>
    <items>
      <rdf:Seq>	<rdf:li rdf:resource="https://www.journals.uchicago.edu/doi/10.1086/724447"/>
	<rdf:li rdf:resource="http://computationalculture.net/situating-bayesian-knowledge/"/>
	<rdf:li rdf:resource="https://www.cambridge.org/core/journals/philosophy-of-science/article/confidence-in-probabilistic-risk-assessment/E7F3CBDE94325C92D5DBFE30A5FAD235?WT.mc_id=New%2520Cambridge%2520Alert%2520-%2520Issues"/>
	<rdf:li rdf:resource="https://www.cambridge.org/core/journals/philosophy-of-science/article/pride-and-probability/CB15DB2546E744E9DCDDB2F4D14633B4?WT.mc_id=New%2520Cambridge%2520Alert%2520-%2520Issues"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2304.06670"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2110.06581"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2402.16422"/>
	<rdf:li rdf:resource="http://approximateinference.org/accepted/"/>
	<rdf:li rdf:resource="https://doi.org/10.2307/3008187"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2305.08429"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2107.04562"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2206.02068"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2205.13698"/>
	<rdf:li rdf:resource="https://osf.io/b8tvk/"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1705.03439"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2107.03584"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2107.02522"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2104.07359"/>
	<rdf:li rdf:resource="http://www.mit.edu/~mitter/publications/100_variational_approach_SIAM.pdf"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2004.06425"/>
	<rdf:li rdf:resource="http://philsci-archive.pitt.edu/18496/"/>
	<rdf:li rdf:resource="http://www.unshieldedcolliders.net/2020/05/belated-reply-to-huemer-on-popper.html"/>
	<rdf:li rdf:resource="https://donskerclass.github.io/post/some-issues-with-bayesian-epistemology/"/>
	<rdf:li rdf:resource="http://philsci-archive.pitt.edu/12429/"/>
	<rdf:li rdf:resource="https://link.springer.com/article/10.1007/s42113-019-00032-3"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1907.03809"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1909.06523"/>
	<rdf:li rdf:resource="https://amstat.tandfonline.com/doi/full/10.1080/10618600.2019.1637749"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1908.00882"/>
	<rdf:li rdf:resource="https://onlinelibrary.wiley.com/doi/abs/10.1111/sjos.12393?af=R"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1905.11448"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1902.00640"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1905.10466"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1502.06045"/>
	<rdf:li rdf:resource="http://alisongopnik.com/Papers_Alison/default.htm"/>
	<rdf:li rdf:resource="http://djnavarro.net/post/2018-09-15-open-closed"/>
	<rdf:li rdf:resource="https://link.springer.com/article/10.1007/s11238-017-9615-y"/>
	<rdf:li rdf:resource="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5805941/"/>
	<rdf:li rdf:resource="http://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-041715-033523"/>
	<rdf:li rdf:resource="http://www.cambridge.org/us/academic/subjects/statistics-probability/statistical-theory-and-methods/fundamentals-nonparametric-bayesian-inference?format=HB#yKtRQ6H8uhqc5zKX.97"/>
	<rdf:li rdf:resource="http://nostalgebraist.tumblr.com/post/161645122124/bayes-a-kinda-sorta-masterpost"/>
	<rdf:li rdf:resource="http://www.jstor.org/stable/1884324?seq=1#page_scan_tab_contents"/>
	<rdf:li rdf:resource="http://www.mitpressjournals.org/doi/abs/10.1162/NECO_a_00912#.WH0fI7GZOL8"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11229-015-0800-7"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1507.01059"/>
	<rdf:li rdf:resource="http://rstl.royalsocietypublishing.org/content/53/370.short"/>
	<rdf:li rdf:resource="http://www.pnas.org/content/112/12/3788.abstract.html?etoc"/>
	<rdf:li rdf:resource="http://pss.sagepub.com/content/24/12/2351"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1411.3984"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1412.3442"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1410.7600"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1501.03326"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1412.3730"/>
	<rdf:li rdf:resource="http://philsci-archive.pitt.edu/3165/"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11229-013-0286-0"/>
	<rdf:li rdf:resource="http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2374040"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1403.4630"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1211.1530"/>
	<rdf:li rdf:resource="http://www.tandfonline.com/doi/abs/10.1198/jcgs.2011.09090#.Uu_-Dv0aAlM"/>
	<rdf:li rdf:resource="http://jmlr.org/papers/v14/fukumizu13a.html"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1312.2974"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1311.0072"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1312.0302"/>
	<rdf:li rdf:resource="http://www.aeaweb.org/articles.php?doi=10.1257/aer.103.7.2790"/>
	<rdf:li rdf:resource="http://www.jstor.org/discover/10.1086/673249"/>
	<rdf:li rdf:resource="https://www-cs.stanford.edu/people/slingamn/philosophy/frequentism_positivism/frequentism_positivism.pdf"/>
	<rdf:li rdf:resource="http://www.jneurosci.org/content/33/26/10887.short?rss=1"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1306.6430"/>
	<rdf:li rdf:resource="http://ba.stat.cmu.edu/abstracts/Scutari.php"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1306.3092"/>
      </rdf:Seq>
    </items>
  </channel><item rdf:about="https://www.journals.uchicago.edu/doi/10.1086/724447">
    <title>On the Ecological and Internal Rationality of Bayesian Conditionalization and Other Belief Updating Strategies | The British Journal for the Philosophy of Science: Vol 77, No 1</title>
    <dc:date>2026-04-18T22:13:36+00:00</dc:date>
    <link>https://www.journals.uchicago.edu/doi/10.1086/724447</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["According to Bayesians, agents should respond to evidence by conditionalizing their prior degrees of belief on what they learn. A major aim of this article is to demonstrate that there are common scenarios where Bayesian conditionalization is less rational—from both an ecological and an internal perspective—than other theoretically well-motivated belief updating strategies, even in simple situations and even for an ‘ideal’ agent who is computationally unbounded. The examples also serve to demarcate the conditions under which Bayesian conditionalization may be expected to be ecologically optimal. A second aim of the article is to argue for a broader notion of rationality than what is typically assumed in formal epistemology. On this broader understanding of rationality, classical decision theoretic principles such as expected utility maximization play a less important role."]]></description>
<dc:subject>to:NB epistemology bayesianism rationality</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c9a42dd78d1c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:epistemology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:rationality"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://computationalculture.net/situating-bayesian-knowledge/">
    <title>Situating Bayesian Knowledge: A Case Study of Modelling Pollutant Transfers from Land to Water – Computational Culture</title>
    <dc:date>2025-07-17T14:36:14+00:00</dc:date>
    <link>http://computationalculture.net/situating-bayesian-knowledge/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian statistics is an alternative to classical, frequentist statistics. Some have argued that the Bayesian framework embodies a necessarily ‘subjective’ perspective which accounts for the context dependence and relativity of knowledge and contrasts it with a frequentist approach. We argue that such epistemological discussions can actually obscure all the ways in which Bayesian knowledge is partial and, in this way, similar to frequentist knowledge. In this contribution, we explore the contingency and performativity of knowing in Bayesian ways by revisiting an application of Bayesian modelling in a case of pollutant transfers from land to water. We query this material from an STS perspective, thinking through a concrete Bayesian modelling process, the various choices made and their alternatives. We ponder how specific practices that play out in Bayesian modelling–model building, data preparation, setting the prior, defining the likelihood function, sampling from the posterior, and checking the model–produce knowledges that can be situated within and produce, for example, partial perspectives on the issue in question, on knowledge and the good, and within social and material contexts. We touch upon discussions on mathematical affordances of Bayesian modelling such as the lack of a built-in mechanism for updating the space of models. Ultimately, we discuss how Bayesian modelling practices enact aspects of the world, including ‘natural’, ‘social’, ‘political’ and ‘ethical’ ‘objects’, and can (re)configure (social) relations. We demonstrate the value of collaborating with actors that can unsettle the Bayesian workflow to iteratively preserve onto-epistemic openings."

--- I may need AEO to help me understand this...]]></description>
<dc:subject>to:NB to_read to_read_maybe the_french_disease bayesianism hydrology spatio-temporal_statistics re:phil-of-bayes_paper</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e7e6be4024ca/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read_maybe"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:the_french_disease"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hydrology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:spatio-temporal_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.cambridge.org/core/journals/philosophy-of-science/article/confidence-in-probabilistic-risk-assessment/E7F3CBDE94325C92D5DBFE30A5FAD235?WT.mc_id=New%2520Cambridge%2520Alert%2520-%2520Issues">
    <title>Confidence in Probabilistic Risk Assessment | Philosophy of Science | Cambridge Core</title>
    <dc:date>2024-10-09T19:47:32+00:00</dc:date>
    <link>https://www.cambridge.org/core/journals/philosophy-of-science/article/confidence-in-probabilistic-risk-assessment/E7F3CBDE94325C92D5DBFE30A5FAD235?WT.mc_id=New%2520Cambridge%2520Alert%2520-%2520Issues</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Epistemic uncertainties are included in probabilistic risk assessment (PRA) as second-order probabilities that represent the degrees of belief of the scientists that a model is correct. In this article, I propose an alternative approach that incorporates the scientist’s confidence in a probability set for a given quantity. First, I give some arguments against the use of precise probabilities to estimate scientific uncertainty in risk analysis. I then extend the “confidence approach” developed by Brian Hill and Richard Bradley to PRA. Finally, I claim that this approach represents model uncertainty better than the standard (Bayesian) model does."]]></description>
<dc:subject>to:NB uncertainty probability bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:bba4da2bc846/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.cambridge.org/core/journals/philosophy-of-science/article/pride-and-probability/CB15DB2546E744E9DCDDB2F4D14633B4?WT.mc_id=New%2520Cambridge%2520Alert%2520-%2520Issues">
    <title>Pride and Probability | Philosophy of Science | Cambridge Core</title>
    <dc:date>2024-10-09T19:46:27+00:00</dc:date>
    <link>https://www.cambridge.org/core/journals/philosophy-of-science/article/pride-and-probability/CB15DB2546E744E9DCDDB2F4D14633B4?WT.mc_id=New%2520Cambridge%2520Alert%2520-%2520Issues</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian agents, argues Belot, are orgulous: they believe in inductive success even when guaranteed to fail on a topologically typical collection of data streams. Here we shed light on how pervasive this phenomenon is. We identify several classes of inductive problems for which Bayesian convergence to the truth is topologically typical. However, we also show that, for all sufficiently complex classes, there are inductive problems for which convergence is topologically atypical. Last, we identify specific topologically typical collections of data streams, observing which guarantees convergence to the truth across all problems from certain natural classes of effective inductive problems."]]></description>
<dc:subject>probability topology bayesianism bayesian_consistency in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:bafe97f64203/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:topology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2304.06670">
    <title>[2304.06670] Do deep neural networks have an inbuilt Occam's razor?</title>
    <dc:date>2024-10-08T22:55:19+00:00</dc:date>
    <link>https://arxiv.org/abs/2304.06670</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The remarkable performance of overparameterized deep neural networks (DNNs) must arise from an interplay between network architecture, training algorithms, and structure in the data. To disentangle these three components, we apply a Bayesian picture, based on the functions expressed by a DNN, to supervised learning. The prior over functions is determined by the network, and is varied by exploiting a transition between ordered and chaotic regimes. For Boolean function classification, we approximate the likelihood using the error spectrum of functions on data. When combined with the prior, this accurately predicts the posterior, measured for DNNs trained with stochastic gradient descent. This analysis reveals that structured data, combined with an intrinsic Occam's razor-like inductive bias towards (Kolmogorov) simple functions that is strong enough to counteract the exponential growth of the number of functions with complexity, is a key to the success of DNNs."

--- I am skeptical from the abstract _alone_.  (Kolmogorov complexity is relative to a choice of universal Turing machine, for starters.)  More broadly, the idea that some form of Occam is automagically granted by (non-crazy) priors is very dubious, because it's vulnerable to essentially the same counter-argument as the one I gave in [http://bactra.org/notebooks/occam-bounds-for-long-programs.html] against the short-programs-generalize-well version of Occam.  Someone who re-arranged his prior to put a lot of probability on a _small_ set of really complex programs/functions would get the same sort of generalization guarantees...  I will, of course, be happy if this paper is actually very insightful and I need to eat all these words after reading it.]]></description>
<dc:subject>via:vaguery learning_theory neural_networks bayesianism to_read</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f2d4bd730be9/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:vaguery"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:learning_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neural_networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2110.06581">
    <title>[2110.06581] A Trust Crisis In Simulation-Based Inference? Your Posterior Approximations Can Be Unfaithful</title>
    <dc:date>2024-08-21T17:35:35+00:00</dc:date>
    <link>https://arxiv.org/abs/2110.06581</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We present extensive empirical evidence showing that current Bayesian simulation-based inference algorithms can produce computationally unfaithful posterior approximations. Our results show that all benchmarked algorithms -- (Sequential) Neural Posterior Estimation, (Sequential) Neural Ratio Estimation, Sequential Neural Likelihood and variants of Approximate Bayesian Computation -- can yield overconfident posterior approximations, which makes them unreliable for scientific use cases and falsificationist inquiry. Failing to address this issue may reduce the range of applicability of simulation-based inference. For this reason, we argue that research efforts should be made towards theoretical and methodological developments of conservative approximate inference algorithms and present research directions towards this objective. In this regard, we show empirical evidence that ensembling posterior surrogates provides more reliable approximations and mitigates the issue."]]></description>
<dc:subject>to_read simulation-based_inference bayesianism in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a523f93a2968/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:simulation-based_inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2402.16422">
    <title>[2402.16422] Bayesian nonparametric statistics, St-Flour lecture notes</title>
    <dc:date>2024-02-27T20:00:22+00:00</dc:date>
    <link>https://arxiv.org/abs/2402.16422</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["These are lecture notes of the 51st Saint-Flour summer school, July 2023, on the topic of Bayesian nonparametric statistics"
--- 186pp]]></description>
<dc:subject>to:NB nonparametrics bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:88ac8b8a01b7/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:nonparametrics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://approximateinference.org/accepted/">
    <title>5th Symposium on Approximate Bayesian Inference (2023)</title>
    <dc:date>2023-12-08T14:10:44+00:00</dc:date>
    <link>http://approximateinference.org/accepted/</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>statistics computational_statistics bayesianism to_download</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:01cc79977083/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_download"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://doi.org/10.2307/3008187">
    <title>A Bayesian Approach to Short-term Forecasting (Harrison and Stevens, 1971)</title>
    <dc:date>2023-09-21T16:50:23+00:00</dc:date>
    <link>https://doi.org/10.2307/3008187</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A new approach to short-term forecasting is described, based on Bayesian principles. The performance of conventional systems is often upset by the occurrence of changes in trend and slope, or transients. In this approach events of this nature are modelled explicitly, and successive data points are used to calculate the posterior probabilities of such events at each instant of time. The system produces not only single-figure forecasts but distributions of trend and slope values which are relevant to subsequent decisions based on forecasts."

--- Apparently what Project Cybersyn was actually trying to implement (per Medina, _Cybernetic Revolutionaries_, p. 267n33).
--- Also, there are at least 3 different DOIs for this paper (JSTOR, Taylor&Francis, Springer/Nature).  Heck of a system we've got here...
--- ETA after reading: cute use of the Kalman filter to do change-point detection, and hazard a guess about the nature of the change.  I think this'd be about one page of R code, mostly to set up the alternatives...]]></description>
<dc:subject>in_NB project_cybersyn time_series prediction change-point_problem non-stationarity bayesianism cybersyn have_read</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:3899e3000a29/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:project_cybersyn"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:time_series"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:change-point_problem"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:non-stationarity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cybersyn"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2305.08429">
    <title>[2305.08429] Bayesian inference for misspecified generative models</title>
    <dc:date>2023-06-08T15:32:21+00:00</dc:date>
    <link>https://arxiv.org/abs/2305.08429</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian inference is a powerful tool for combining information in complex settings, a task of increasing importance in modern applications. However, Bayesian inference with a flawed model can produce unreliable conclusions. This review discusses approaches to performing Bayesian inference when the model is misspecified, where by misspecified we mean that the analyst is unwilling to act as if the model is correct. Much has been written about this topic, and in most cases we do not believe that a conventional Bayesian analysis is meaningful when there is serious model misspecification. Nevertheless, in some cases it is possible to use a well-specified model to give meaning to a Bayesian analysis of a misspecified model and we will focus on such cases. Three main classes of methods are discussed - restricted likelihood methods, which use a model based on a non-sufficient summary of the original data; modular inference methods which use a model constructed from coupled submodels and some of the submodels are correctly specified; and the use of a reference model to construct a projected posterior or predictive distribution for a simplified model considered to be useful for prediction or interpretation."]]></description>
<dc:subject>to:NB bayesianism misspecification statistics re:phil-of-bayes_paper to_read</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:683fd1cb974b/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2107.04562">
    <title>[2107.04562] The Bayesian Learning Rule</title>
    <dc:date>2023-03-18T14:30:57+00:00</dc:date>
    <link>https://arxiv.org/abs/2107.04562</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We show that many machine-learning algorithms are specific instances of a single algorithm called the Bayesian learning rule. The rule, derived from Bayesian principles, yields a wide-range of algorithms from fields such as optimization, deep learning, and graphical models. This includes classical algorithms such as ridge regression, Newton's method, and Kalman filter, as well as modern deep-learning algorithms such as stochastic-gradient descent, RMSprop, and Dropout. The key idea in deriving such algorithms is to approximate the posterior using candidate distributions estimated by using natural gradients. Different candidate distributions result in different algorithms and further approximations to natural gradients give rise to variants of those algorithms. Our work not only unifies, generalizes, and improves existing algorithms, but also helps us design new ones."]]></description>
<dc:subject>to:NB learning_theory optimization bayesianism color_me_skeptical have_skimmed</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:928b90fa32e5/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:learning_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:color_me_skeptical"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_skimmed"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2206.02068">
    <title>[2206.02068] A Further Look at the Bayes Blind Spot</title>
    <dc:date>2022-06-09T08:37:09+00:00</dc:date>
    <link>https://arxiv.org/abs/2206.02068</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Gyenis and Redei have demonstrated that any prior p on a finite algebra, however chosen, severely restricts the set of posteriors accessible from p by Jeffrey conditioning on a nontrivial partition. Their demonstration involves showing that the set of posteriors not accessible from p in this way (which they call the Bayes blind spot of p) is large with respect to three common measures of size, namely, having cardinality c, (normalized) Lebesgue measure 1, and Baire second category with respect to a natural topology. In the present paper, we establish analogous results for probability measures defined on any infinite sigma algebra of subsets of a denumerably infinite set. However, we have needed to employ distinctly different approaches to determine the cardinality, and especially, the topological and measure-theoretic sizes of the Bayes blind spot in the infinite case. Interestingly, all of the results that we establish for a single prior p continue to hold for the intersection of the Bayes blind spots of countably many priors. This leads us to conjecture that Bayesian learning itself might be just as culpable as the limitations imposed by priors in enabling the existence of large Bayes blind spots."]]></description>
<dc:subject>to:NB bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:3845d5d84905/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2205.13698">
    <title>[2205.13698] Characterizing the robustness of Bayesian adaptive experimental designs to active learning bias</title>
    <dc:date>2022-05-30T20:25:32+00:00</dc:date>
    <link>https://arxiv.org/abs/2205.13698</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian adaptive experimental design is a form of active learning, which chooses samples to maximize the information they give about uncertain parameters. Prior work has shown that other forms of active learning can suffer from active learning bias, where unrepresentative sampling leads to inconsistent parameter estimates. We show that active learning bias can also afflict Bayesian adaptive experimental design, depending on model misspecification. We develop an information-theoretic measure of misspecification, and show that worse misspecification implies more severe active learning bias. At the same time, model classes incorporating more "noise" - i.e., specifying higher inherent variance in observations - suffer less from active learning bias, because their predictive distributions are likely to overlap more with the true distribution. Finally, we show how these insights apply to a (simulated) preference learning experiment."]]></description>
<dc:subject>bayesianism experimental_design psychology self-promotion misspecification</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:b9728939bfda/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:experimental_design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:self-promotion"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://osf.io/b8tvk/">
    <title>OSF Preprints | Directional motives and different priors are observationally equivalent</title>
    <dc:date>2021-12-13T06:21:52+00:00</dc:date>
    <link>https://osf.io/b8tvk/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Many experimental and observational studies use the way that subjects respond to information as evidence that partisan bias or directional motives influence (or do not influence) political beliefs. For a natural and tractable formulation belief formation with both accuracy and directional motives, this is not possible. Any subject influenced by directional motives has a "Fully Bayesian Equivalent" with identical beliefs upon observing any signal. As a result, comparing how individuals or groups with different partisanship or priors respond to information has no diagnostic value in detecting motivated reasoning, even in a multivariate or dynamic setting. Conversely, providing a ``Bayesian rationalization'' consistent with a pattern of updating is not meaningful evidence for a lack of directional motives. These results have theoretical implications for the convergence of beliefs among those with directional motives and practical implications for empirical studies that aim to detect directional motives."]]></description>
<dc:subject>psychology bounded_rationality bayesianism to:blog have_read in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:bced20d10754/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bounded_rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1705.03439">
    <title>[1705.03439] Frequentist Consistency of Variational Bayes</title>
    <dc:date>2021-07-11T05:56:27+00:00</dc:date>
    <link>https://arxiv.org/abs/1705.03439</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A key challenge for modern Bayesian statistics is how to perform scalable inference of posterior distributions. To address this challenge, variational Bayes (VB) methods have emerged as a popular alternative to the classical Markov chain Monte Carlo (MCMC) methods. VB methods tend to be faster while achieving comparable predictive performance. However, there are few theoretical results around VB. In this paper, we establish frequentist consistency and asymptotic normality of VB methods. Specifically, we connect VB methods to point estimates based on variational approximations, called frequentist variational approximations, and we use the connection to prove a variational Bernstein-von Mises theorem. The theorem leverages the theoretical characterizations of frequentist variational approximations to understand asymptotic properties of VB. In summary, we prove that (1) the VB posterior converges to the Kullback-Leibler (KL) minimizer of a normal distribution, centered at the truth and (2) the corresponding variational expectation of the parameter is consistent and asymptotically normal. As applications of the theorem, we derive asymptotic properties of VB posteriors in Bayesian mixture models, Bayesian generalized linear mixed models, and Bayesian stochastic block models. We conduct a simulation study to illustrate these theoretical results."]]></description>
<dc:subject>to:NB computational_statistics bayesianism bayesian_consistency blei.david statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f8a84a82e174/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:blei.david"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2107.03584">
    <title>[2107.03584] Evaluating Sensitivity to the Stick-Breaking Prior in Bayesian Nonparametrics</title>
    <dc:date>2021-07-11T05:54:20+00:00</dc:date>
    <link>https://arxiv.org/abs/2107.03584</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian models based on the Dirichlet process and other stick-breaking priors have been proposed as core ingredients for clustering, topic modeling, and other unsupervised learning tasks. Prior specification is, however, relatively difficult for such models, given that their flexibility implies that the consequences of prior choices are often relatively opaque. Moreover, these choices can have a substantial effect on posterior inferences. Thus, considerations of robustness need to go hand in hand with nonparametric modeling. In the current paper, we tackle this challenge by exploiting the fact that variational Bayesian methods, in addition to having computational advantages in fitting complex nonparametric models, also yield sensitivities with respect to parametric and nonparametric aspects of Bayesian models. In particular, we demonstrate how to assess the sensitivity of conclusions to the choice of concentration parameter and stick-breaking distribution for inferences under Dirichlet process mixtures and related mixture models. We provide both theoretical and empirical support for our variational approach to Bayesian sensitivity analysis."]]></description>
<dc:subject>to:NB bayesianism statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:ffd9f867f14c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2107.02522">
    <title>[2107.02522] A nonBayesian view of Hempel's paradox of the ravens</title>
    <dc:date>2021-07-08T16:29:56+00:00</dc:date>
    <link>https://arxiv.org/abs/2107.02522</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In Hempel's paradox of the ravens, seeing a red pencil is considered as supporting evidence that all ravens are black. Also known as the Paradox of Confirmation, the paradox and its many resolutions indicate that we cannot underestimate the logical and statistical elements needed in the assessment of evidence in support of a hypothesis. Most of the previous analyses of the paradox are within the Bayesian framework. These analyses and Hempel himself generally accept the paradoxical conclusion; it feels paradoxical supposedly because the amount of evidence is extremely small. Here I describe a nonBayesian analysis of various statistical models with an accompanying likelihood-based reasoning. The analysis shows that the paradox seems paradoxical because there are natural models where observing a red pencil has no relevance to the color of ravens. In general the value of the evidence depends crucially on the sampling scheme and on the assumption about the underlying parameters of the relevant model."

--- IIRC this is also what Peter Godfrey-Smith says in _Theory and Reality_.]]></description>
<dc:subject>to:NB epistemology bayesianism likelihood</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:531825dfad91/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:epistemology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:likelihood"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2104.07359">
    <title>[2104.07359] Robust Generalised Bayesian Inference for Intractable Likelihoods</title>
    <dc:date>2021-04-16T19:39:01+00:00</dc:date>
    <link>https://arxiv.org/abs/2104.07359</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Generalised Bayesian inference updates prior beliefs using a loss function, rather than a likelihood, and can therefore be used to confer robustness against possible misspecification of the likelihood. Here we consider generalised Bayesian inference with a Stein discrepancy as a loss function, motivated by applications in which the likelihood contains an intractable normalisation constant. In this context, the Stein discrepancy circumvents evaluation of the normalisation constant and produces generalised posteriors that are either closed form or accessible using standard Markov chain Monte Carlo. On a theoretical level, we show consistency, asymptotic normality, and bias-robustness of the generalised posterior, highlighting how these properties are impacted by the choice of Stein discrepancy. Then, we provide numerical experiments on a range of intractable distributions, including applications to kernel-based exponential family models and non-Gaussian graphical models."

--- If instead of selectively breeding parameter values to have low Kullback divergence, we selectively breed parameter values to minimize another discrepancy, selection will still converge on the fitness-maximizer(s).]]></description>
<dc:subject>to:NB bayesianism bayesian_consistency re:HEAS likelihood re:bayes_as_evol</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:ff2c020c49a1/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:HEAS"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:likelihood"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:bayes_as_evol"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.mit.edu/~mitter/publications/100_variational_approach_SIAM.pdf">
    <title>A Variational Approach to Nonlinear Estimation</title>
    <dc:date>2020-12-15T15:12:36+00:00</dc:date>
    <link>http://www.mit.edu/~mitter/publications/100_variational_approach_SIAM.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We consider estimation problems, in which the estimand, X, and observation, Y ,
take values in measurable spaces. Regular conditional versions of the forward and inverse Bayes
formula are shown to have dual variational characterizations involving the minimization of apparent
information and the maximization of compatible information. These both have natural informationtheoretic interpretations, according to which Bayes’ formula and its inverse are optimal information
processors. The variational characterization of the forward formula has the same form as that of Gibbs
measures in statistical mechanics. The special case in which X and Y are diffusion processes governed
by stochastic differential equations is examined in detail. The minimization of apparent information
can then be formulated as a stochastic optimal control problem, with cost that is quadratic in both
the control and observation fit. The dual problem can be formulated in terms of infinite-dimensional
deterministic optimal control. Local versions of the variational characterizations are developed which
quantify information flow in the estimators. In this context, the information conserving property of
Bayesian estimators coincides with the Davis–Varaiya martingale stochastic dynamic programming
principle."]]></description>
<dc:subject>to:NB to_read filtering bayesianism via:mraginsky</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c6f424e7e845/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:filtering"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:mraginsky"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2004.06425">
    <title>[2004.06425] Computing Bayes: Bayesian Computation from 1763 to the 21st Century</title>
    <dc:date>2020-12-14T14:57:35+00:00</dc:date>
    <link>https://arxiv.org/abs/2004.06425</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The Bayesian statistical paradigm uses the language of probability to express uncertainty about the phenomena that generate observed data. Probability distributions thus characterize Bayesian analysis, with the rules of probability used to transform prior probability distributions for all unknowns - parameters, latent variables, models - into posterior distributions, subsequent to the observation of data. Conducting Bayesian analysis requires the evaluation of integrals in which these probability distributions appear. Bayesian computation is all about evaluating such integrals in the typical case where no analytical solution exists. This paper takes the reader on a chronological tour of Bayesian computation over the past two and a half centuries. Beginning with the one-dimensional integral first confronted by Bayes in 1763, through to recent problems in which the unknowns number in the millions, we place all computational problems into a common framework, and describe all computational methods using a common notation. The aim is to help new researchers in particular - and more generally those interested in adopting a Bayesian approach to empirical work - make sense of the plethora of computational techniques that are now on offer; understand when and why different methods are useful; and see the links that do exist, between them all."]]></description>
<dc:subject>to:NB bayesianism history_of_statistics computational_statistics robert.christian_p. to_read to_teach:statcomp</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:07e2869f2a9c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:history_of_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:robert.christian_p."/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_teach:statcomp"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://philsci-archive.pitt.edu/18496/">
    <title>Collectivist Foundations for Bayesian Statistics - PhilSci-Archive</title>
    <dc:date>2020-12-12T16:04:04+00:00</dc:date>
    <link>http://philsci-archive.pitt.edu/18496/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["What (if anything) justifies the use of Bayesian statistics in science? The traditional answer is that Bayesian statistics is simply an instance of orthodox expected utility theory. Thus, Bayesian statistical methods, like principles of utility theory, are justified by norms of individual rationality. In particular, most Bayesians argue that a scientist's credences must satisfy the probability axioms if she adheres to norms of practical and epistemic (individual) rationality. We argue that, to justify Bayesian statistics as a tool for science, it is necessary that a scientist's public credences (i.e., her degrees of belief qua scientist) obey the probability axioms. We claim that norms of collective science help justify this restricted view, termed public probabilism."]]></description>
<dc:subject>to:NB bayesianism philosophy_of_science science_as_a_social_process mayo-wilson.conor re:phil-of-bayes_paper</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:09130a3e5710/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:science_as_a_social_process"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:mayo-wilson.conor"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.unshieldedcolliders.net/2020/05/belated-reply-to-huemer-on-popper.html">
    <title>Belated Reply to Huemer on Popper | Unshielded Colliders</title>
    <dc:date>2020-11-30T06:43:08+00:00</dc:date>
    <link>http://www.unshieldedcolliders.net/2020/05/belated-reply-to-huemer-on-popper.html</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>philosophy_of_science popper.karl bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a07be5854687/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:popper.karl"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://donskerclass.github.io/post/some-issues-with-bayesian-epistemology/">
    <title>Some issues with Bayesian epistemology | David Childers</title>
    <dc:date>2020-11-02T01:08:40+00:00</dc:date>
    <link>https://donskerclass.github.io/post/some-issues-with-bayesian-epistemology/</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>bayesianism kith_and_kin to:blog</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:5fbce078233d/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kith_and_kin"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://philsci-archive.pitt.edu/12429/">
    <title>Solomonoff Prediction and Occam's Razor - PhilSci-Archive</title>
    <dc:date>2020-08-04T19:44:56+00:00</dc:date>
    <link>http://philsci-archive.pitt.edu/12429/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Algorithmic information theory gives an idealized notion of compressibility, that is often presented as an objective measure of simplicity. It is suggested at times that Solomonoff prediction, or algorithmic information theory in a predictive setting, can deliver an argument to justify Occam's razor. This paper explicates the relevant argument, and, by converting it into a Bayesian framework, reveals why it has no such justificatory force. The supposed simplicity concept is better perceived as a specific inductive assumption, the assumption of effectiveness. It is this assumption that is the characterizing element of Solomonoff prediction, and wherein its philosophical interest lies."

--- On skimming: the predictive power of Solomonoff prediction is just a special case of Bayesian consistency, assuming the prior is well-specified.

--- On reading: the predictive power of Solomonoff prediction is just a special case of Bayesian predictors having low regret bounds.  ("Low" in the sense of growth rate with the length of the data stream; it can be arbitrarily large depending on how little prior probability the retrospectively-optimal distribution was given.)  Any prior over effectively-computable (semi-) measures is the same as doing Solmonoff prediction for some universal computer, so this really is just about a Bayesian assuming their prior (that Nature is effectively-computable) is well-specified.]]></description>
<dc:subject>algorithmic_information_theory prediction induction via:csantos bayesianism bayesian_consistency have_read low-regret_learning in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a0ab9c283178/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:algorithmic_information_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:induction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:csantos"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:low-regret_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://link.springer.com/article/10.1007/s42113-019-00032-3">
    <title>Computational Resource Demands of a Predictive Bayesian Brain | SpringerLink</title>
    <dc:date>2020-01-06T18:29:49+00:00</dc:date>
    <link>https://link.springer.com/article/10.1007/s42113-019-00032-3</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["There is a growing body of evidence that the human brain may be organized according to principles of predictive processing. An important conjecture in neuroscience is that a brain organized in this way can effectively and efficiently approximate Bayesian inferences. Given that many forms of cognition seem to be well characterized as a form of Bayesian inference, this conjecture has great import for cognitive science. It suggests that predictive processing may provide a neurally plausible account of how forms of cognition that are modeled as Bayesian inference may be physically implemented in the brain. Yet, as we show in this paper, the jury is still out on whether or not the conjecture is really true. Specifically, we demonstrate that each key subcomputation invoked in predictive processing potentially hides a computationally intractable problem. We discuss the implications of these sobering results for the predictive processing account and propose a way to move forward."]]></description>
<dc:subject>to:NB neuroscience neural_coding_and_decoding perception cognitive_science bayesianism computational_complexity</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a8142c60720b/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neuroscience"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neural_coding_and_decoding"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:perception"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_complexity"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1907.03809">
    <title>[1907.03809] Competing Models</title>
    <dc:date>2019-10-22T13:42:08+00:00</dc:date>
    <link>https://arxiv.org/abs/1907.03809</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We develop a model in which different agents compete to predict a variable of interest. This variable is related to observables via an unknown data generating process. All agents are Bayesian, but may have `misspecified models' of the world, i.e., they consider different subsets of observables to make their prediction. After observing a common dataset, who has the highest confidence in her predictive ability? We characterize it and show that it crucially depends on the size of the dataset. With big data, we show it is typically `large-dimensional,' possibly using more variables than the true model. With small data, we show (under additional assumptions) that it is an agent using a model that is `small-dimensional,' in the sense of considering fewer covariates than the true data generating process. The theory is applied to auctions of assets where bidders observe the same information but hold different priors."

--- I'm a bit puzzled by the abstract, since it's trivial that for any Bayesian agent, $\Pr_{me}(My model is correct|Data)=1$, regardless of the data.  Every Bayesian agent begins and remains 100% confident that the truth is in the support of their prior.]]></description>
<dc:subject>prediction bayesianism model_selection in_NB color_me_skeptical</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a6ec7398835b/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:model_selection"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:color_me_skeptical"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1909.06523">
    <title>[1909.06523] Justifying the Norms of Inductive Inference</title>
    <dc:date>2019-09-18T12:55:38+00:00</dc:date>
    <link>https://arxiv.org/abs/1909.06523</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian inference is limited in scope because it cannot be applied in idealized contexts where none of the hypotheses under consideration is true and because it is committed to always using the likelihood as a measure of evidential favoring, even when that is inappropriate. The purpose of this paper is to study inductive inference in a very general setting where finding the truth is not necessarily the goal and where the measure of evidential favoring is not necessarily the likelihood. I use an accuracy argument to argue for probabilism and I develop a new kind of argument to argue for two general updating rules, both of which are reasonable in different contexts. One of the updating rules has standard Bayesian updating, Bissiri et al's (2016) general Bayesian updating, Douven's (2016) IBE-based updating, and Vassend's (2019a) quasi-Bayesian updating as special cases. The other updating rule is novel."]]></description>
<dc:subject>to:NB bayesianism philosophy_of_science re:phil-of-bayes_paper color_me_skeptical</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a5839291fd59/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:color_me_skeptical"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://amstat.tandfonline.com/doi/full/10.1080/10618600.2019.1637749">
    <title>Testing Sparsity-Inducing Penalties: Journal of Computational and Graphical Statistics: Vol 0, No 0</title>
    <dc:date>2019-08-20T16:07:38+00:00</dc:date>
    <link>https://amstat.tandfonline.com/doi/full/10.1080/10618600.2019.1637749</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Many penalized maximum likelihood estimators correspond to posterior mode estimators under specific prior distributions. Appropriateness of a particular class of penalty functions can therefore be interpreted as the appropriateness of a prior for the parameters. For example, the appropriateness of a lasso penalty for regression coefficients depends on the extent to which the empirical distribution of the regression coefficients resembles a Laplace distribution. We give a testing procedure of whether or not a Laplace prior is appropriate and accordingly, whether or not using a lasso penalized estimate is appropriate. This testing procedure is designed to have power against exponential power priors which correspond to ℓqℓq penalties. Via simulations, we show that this testing procedure achieves the desired level and has enough power to detect violations of the Laplace assumption when the numbers of observations and unknown regression coefficients are large. We then introduce an adaptive procedure that chooses a more appropriate prior and corresponding penalty from the class of exponential power priors when the null hypothesis is rejected. We show that this can improve estimation of the regression coefficients both when they are drawn from an exponential power distribution and when they are drawn from a spike-and-slab distribution. Supplementary materials for this article are available online."

--- I feel like I fundamentally disagree with this approach.  Those priors are merely (to quote Jamie Robins and Larry Wasserman) "frequentist pursuit", and have no bearing on whether (say) the Lasso will give a good sparse, linear approximation to the underlying regression function (see https://normaldeviate.wordpress.com/2013/09/11/consistency-sparsistency-and-presistency/).  All of which said, Hoff is always worth listening to, so the last tag applies with special force.]]></description>
<dc:subject>to:NB model_checking sparsity regression hypothesis_testing bayesianism re:phil-of-bayes_paper hoff.peter to_besh</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:aaba8d8a838f/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:model_checking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:sparsity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:regression"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hypothesis_testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hoff.peter"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_besh"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1908.00882">
    <title>[1908.00882] Population Predictive Checks</title>
    <dc:date>2019-08-05T12:47:41+00:00</dc:date>
    <link>https://arxiv.org/abs/1908.00882</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian modeling has become a staple for researchers analyzing data. Thanks to recent developments in approximate posterior inference, modern researchers can easily build, use, and revise complicated Bayesian models for large and rich data. These new abilities, however, bring into focus the problem of model assessment. Researchers need tools to diagnose the fitness of their models, to understand where a model falls short, and to guide its revision. In this paper we develop a new method for Bayesian model checking, the population predictive check (Pop-PC). Pop-PCs are built on posterior predictive checks (PPC), a seminal method that checks a model by assessing the posterior predictive distribution on the observed data. Though powerful, PPCs use the data twice---both to calculate the posterior predictive and to evaluate it---which can lead to overconfident assessments. Pop-PCs, in contrast, compare the posterior predictive distribution to the population distribution of the data. This strategy blends Bayesian modeling with frequentist assessment, leading to a robust check that validates the model on its generalization. Of course the population distribution is not usually available; thus we use tools like the bootstrap and cross validation to estimate the Pop-PC. Further, we extend Pop-PCs to hierarchical models. We study Pop-PCs on classical regression and a hierarchical model of text. We show that Pop-PCs are robust to overfitting and can be easily deployed on a broad family of models."
]]></description>
<dc:subject>to:NB model_checking bayesianism statistics blei.david re:phil-of-bayes_paper to_read cross-validation</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:221ff74c92f3/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:model_checking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:blei.david"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cross-validation"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://onlinelibrary.wiley.com/doi/abs/10.1111/sjos.12393?af=R">
    <title>Alternatives to post‐processing posterior predictive p values - Gåsemyr - - Scandinavian Journal of Statistics - Wiley Online Library</title>
    <dc:date>2019-06-15T16:54:04+00:00</dc:date>
    <link>https://onlinelibrary.wiley.com/doi/abs/10.1111/sjos.12393?af=R</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The posterior predictive p value (ppp) was invented as a Bayesian counterpart to classical p values. The methodology can be applied to discrepancy measures involving both data and parameters and can, hence, be targeted to check for various modeling assumptions. The interpretation can, however, be difficult since the distribution of the ppp value under modeling assumptions varies substantially between cases. A calibration procedure has been suggested, treating the ppp value as a test statistic in a prior predictive test. In this paper, we suggest that a prior predictive test may instead be based on the expected posterior discrepancy, which is somewhat simpler, both conceptually and computationally. Since both these methods require the simulation of a large posterior parameter sample for each of an equally large prior predictive data sample, we furthermore suggest to look for ways to match the given discrepancy by a computation‐saving conflict measure. This approach is also based on simulations but only requires sampling from two different distributions representing two contrasting information sources about a model parameter. The conflict measure methodology is also more flexible in that it handles non‐informative priors without difficulty. We compare the different approaches theoretically in some simple models and in a more complex applied example."]]></description>
<dc:subject>to:NB bayesianism statistics model_checking re:phil-of-bayes_paper</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:0616078fac6e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:model_checking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1905.11448">
    <title>[1905.11448] Probabilistic mappings and Bayesian nonparametrics</title>
    <dc:date>2019-05-29T21:15:39+00:00</dc:date>
    <link>https://arxiv.org/abs/1905.11448</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In this paper we develop a functorial language of probabilistic mappings and apply it to basic problems in Bayesian nonparametrics. First we extend and unify the Kleisli category of probabilistic mappings proposed by Lawvere and Giry with the category of statistical models proposed by Chentsov and Morse-Sacksteder. Then we introduce the notion of a Bayesian statistical model that formalizes the notion of a parameter space with a given prior distribution in Bayesian statistics. We give a formula for posterior distributions, assuming that the underlying parameter space of a Bayesian statistical model is a Souslin space and the sample space of the Bayesian statistical model is a subset in a complete connected finite dimensional Riemannian manifold. Then we give a new proof of the existence of Dirichlet measures over any measurable space using a functorial property of the Dirichlet map constructed by Sethuraman."]]></description>
<dc:subject>to:NB probability category_theory bayesianism kith_and_kin jost.jurgen statistics barely-comprehensible_mathematics because_everything_is_clearer_with_category_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:4f49fbb81609/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:category_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kith_and_kin"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:jost.jurgen"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:barely-comprehensible_mathematics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:because_everything_is_clearer_with_category_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1902.00640">
    <title>[1902.00640] Particle Flow Bayes' Rule</title>
    <dc:date>2019-05-29T21:14:02+00:00</dc:date>
    <link>https://arxiv.org/abs/1902.00640</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We present a particle flow realization of Bayes' rule, where an ODE-based neural operator is used to transport particles from a prior to its posterior after a new observation. We prove that such an ODE operator exists. Its neural parameterization can be trained in a meta-learning framework, allowing this operator to reason about the effect of an individual observation on the posterior, and thus generalize across different priors, observations and to sequential Bayesian inference. We demonstrated the generalization ability of our particle flow Bayes operator in several canonical and high dimensional examples."]]></description>
<dc:subject>to:NB bayesianism neural_networks computational_statistics statistics re:fitness_sampling</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e8d028fe3724/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neural_networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:fitness_sampling"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1905.10466">
    <title>[1905.10466] Decentralized Bayesian Learning over Graphs</title>
    <dc:date>2019-05-28T17:36:09+00:00</dc:date>
    <link>https://arxiv.org/abs/1905.10466</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We propose a decentralized learning algorithm over a general social network. The algorithm leaves the training data distributed on the mobile devices while utilizing a peer to peer model aggregation method. The proposed algorithm allows agents with local data to learn a shared model explaining the global training data in a decentralized fashion. The proposed algorithm can be viewed as a Bayesian and peer-to-peer variant of federated learning in which each agent keeps a "posterior probability distribution" over a global model parameters. The agent update its "posterior" based on 1) the local training data and 2) the asynchronous communication and model aggregation with their 1-hop neighbors. This Bayesian formulation allows for a systematic treatment of model aggregation over any arbitrary connected graph. Furthermore, it provides strong analytic guarantees on converge in the realizable case as well as a closed form characterization of the rate of convergence. We also show that our methodology can be combined with efficient Bayesian inference techniques to train Bayesian neural networks in a decentralized manner. By empirical studies we show that our theoretical analysis can guide the design of network/social interactions and data partitioning to achieve convergence."


--- But this isn't the properly Bayesian thing to do in this situation!!! Each node would need to have a likelihood for the messages it gets from its neighbors as a function of the global model parameters at each round of updating.  This in turn should reflect the distribution over node-level observations as coarsened by what the neighbors will report.  Clearly, this will be a mess.  What they're proposing is drastically more _tractable_ than what a Bayesian ought to do, but then why insist on a pseudo-Bayesian approach?

--- After reading the author list: Tara knows all that, so I presume there's a good reason (e.g., she's studied non-pseudo-Bayesian approaches, if not to death, then at least to exhaustion).]]></description>
<dc:subject>to:NB distributed_systems collective_cognition bayesianism statistics re:democratic_cognition javidi.tara social_learning</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:3fca8f8ac61b/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:distributed_systems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:collective_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:democratic_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:javidi.tara"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:social_learning"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1502.06045">
    <title>[1502.06045] Model specification via sequential coherence and backward induction</title>
    <dc:date>2019-05-25T22:37:51+00:00</dc:date>
    <link>https://arxiv.org/abs/1502.06045</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This paper describes how to specify probability models for data analysis via a backward induction procedure. The new approach yields coherent, prior-free uncertainty assessment. After presenting some intuition-building examples, the new approach is applied to a kernel density estimator, which leads to a novel method for computing point-wise credible intervals in nonparametric density estimation. The new approach has two additional advantages; 1) the posterior mean density can be accurately approximated without resorting to Monte Carlo simulation and 2) concentration bounds are easily established as a function of sample size."]]></description>
<dc:subject>to:NB bayesianism bayesian_consistency statistics hahn.p._richard</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:6a359015d43a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hahn.p._richard"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://alisongopnik.com/Papers_Alison/default.htm">
    <title>Alison Gopnik Papers</title>
    <dc:date>2018-10-11T15:39:31+00:00</dc:date>
    <link>http://alisongopnik.com/Papers_Alison/default.htm</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>psychology cognitive_development causal_inference bayesianism hume cognitive_science causal_discovery</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e4a1c6e4bf68/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_development"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:causal_inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hume"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:causal_discovery"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://djnavarro.net/post/2018-09-15-open-closed">
    <title>A personal essay on Bayes factors</title>
    <dc:date>2018-09-19T19:08:49+00:00</dc:date>
    <link>http://djnavarro.net/post/2018-09-15-open-closed</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[I would have said nobody blogs like this anymore, and I am very happy to be very wrong.]]></description>
<dc:subject>have_read model_selection bayesianism statistics psychology social_science_methodology via:tslumley</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a62c43579d69/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:model_selection"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:social_science_methodology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:tslumley"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://link.springer.com/article/10.1007/s11238-017-9615-y">
    <title>Learning from others: conditioning versus averaging | SpringerLink</title>
    <dc:date>2018-07-02T14:56:39+00:00</dc:date>
    <link>https://link.springer.com/article/10.1007/s11238-017-9615-y</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["How should we revise our beliefs in response to the expressed probabilistic opinions of experts on some proposition when these experts are in disagreement? In this paper I examine the suggestion that in such circumstances we should adopt a linear average of the experts’ opinions and consider whether such a belief revision policy is compatible with Bayesian conditionalisation. By looking at situations in which full or partial deference to the expressed opinions of others is required by Bayesianism I show that only in trivial circumstances are the requirements imposed by linear averaging compatible with it."

--- This is completely unsurprising, but good to check.]]></description>
<dc:subject>to:NB collective_cognition bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:af8b4f29a7fb/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:collective_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5805941/">
    <title>Trust your gut: using physiological states as a source of information is almost as effective as optimal Bayesian learning</title>
    <dc:date>2018-05-01T20:33:53+00:00</dc:date>
    <link>https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5805941/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Approaches to understanding adaptive behaviour often assume that animals have perfect information about environmental conditions or are capable of sophisticated learning. If such learning abilities are costly, however, natural selection will favour simpler mechanisms for controlling behaviour when faced with uncertain conditions. Here, we show that, in a foraging context, a strategy based only on current energy reserves often performs almost as well as a Bayesian learning strategy that integrates all previous experiences to form an optimal estimate of environmental conditions. We find that Bayesian learning gives a strong advantage only if fluctuations in the food supply are very strong and reasonably frequent. The performance of both the Bayesian and the reserve-based strategy are more robust to inaccurate knowledge of the temporal pattern of environmental conditions than a strategy that has perfect knowledge about current conditions. Studies assuming Bayesian learning are often accused of being unrealistic; our results suggest that animals can achieve a similar level of performance to Bayesians using much simpler mechanisms based on their physiological state. More broadly, our work suggests that the ability to use internal states as a source of information about recent environmental conditions will have weakened selection for sophisticated learning and decision-making systems."

--- Slightly astonishing to see only one reference to Gigerenzer...]]></description>
<dc:subject>to:NB psychology ethology adaptive_behavior bayesianism heuristics via:?</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:85a86c06a9db/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:ethology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:adaptive_behavior"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:heuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:?"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-041715-033523">
    <title>On the Frequentist Properties of Bayesian Nonparametric Methods | Annual Review of Statistics and Its Application</title>
    <dc:date>2017-09-25T13:44:18+00:00</dc:date>
    <link>http://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-041715-033523</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In this paper, I review the main results on the asymptotic properties of the posterior distribution in nonparametric or high-dimensional models. In particular, I explain how posterior concentration rates can be derived and what we learn from such analysis in terms of the impact of the prior distribution on high-dimensional models. These results concern fully Bayes and empirical Bayes procedures. I also describe some of the results that have been obtained recently in semiparametric models, focusing mainly on the Bernstein–von Mises property. Although these results are theoretical in nature, they shed light on some subtle behaviors of the prior models and sharpen our understanding of the family of functionals that can be well estimated for a given prior model."]]></description>
<dc:subject>to:NB bayesian_consistency bayesianism statistics re:bayes_as_evol</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:41b208fba8b4/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:bayes_as_evol"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.cambridge.org/us/academic/subjects/statistics-probability/statistical-theory-and-methods/fundamentals-nonparametric-bayesian-inference?format=HB#yKtRQ6H8uhqc5zKX.97">
    <title>Fundamentals nonparametric bayesian inference | Statistical theory and methods | Cambridge University Press</title>
    <dc:date>2017-08-26T00:33:15+00:00</dc:date>
    <link>http://www.cambridge.org/us/academic/subjects/statistics-probability/statistical-theory-and-methods/fundamentals-nonparametric-bayesian-inference?format=HB#yKtRQ6H8uhqc5zKX.97</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Explosive growth in computing power has made Bayesian methods for infinite-dimensional models - Bayesian nonparametrics - a nearly universal framework for inference, finding practical use in numerous subject areas. Written by leading researchers, this authoritative text draws on theoretical advances of the past twenty years to synthesize all aspects of Bayesian nonparametrics, from prior construction to computation and large sample behavior of posteriors. Because understanding the behavior of posteriors is critical to selecting priors that work, the large sample theory is developed systematically, illustrated by various examples of model and prior combinations. Precise sufficient conditions are given, with complete proofs, that ensure desirable posterior properties and behavior. Each chapter ends with historical notes and numerous exercises to deepen and consolidate the reader's understanding, making the book valuable for both graduate students and researchers in statistics and machine learning, as well as in application areas such as econometrics and biostatistics."

--- These are two of the central figures in the field, and van der Vaart is a really good writer...]]></description>
<dc:subject>to:NB books:noted coveted nonparametrics bayesianism bayesian_consistency statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:ab91c474e1c0/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:books:noted"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:coveted"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:nonparametrics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://nostalgebraist.tumblr.com/post/161645122124/bayes-a-kinda-sorta-masterpost">
    <title>trees are harlequins, words are harlequins — bayes: a kinda-sorta masterpost</title>
    <dc:date>2017-08-13T16:06:12+00:00</dc:date>
    <link>http://nostalgebraist.tumblr.com/post/161645122124/bayes-a-kinda-sorta-masterpost</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[(Last tag is for the cultists whom the poster is [more or less explicitly] going after)]]></description>
<dc:subject>bayesianism statistics foundations_of_statistics utter_stupidity re:phil-of-bayes_paper</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:497d84bc939b/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:utter_stupidity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.jstor.org/stable/1884324?seq=1#page_scan_tab_contents">
    <title>Risk, Ambiguity, and the Savage Axioms on JSTOR</title>
    <dc:date>2017-03-21T04:15:14+00:00</dc:date>
    <link>http://www.jstor.org/stable/1884324?seq=1#page_scan_tab_contents</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["I. Are there uncertainties that are not risks? 643.--II. Uncertainties that are not risks, 647.--III. Why are some uncertainties not risks?--656."

--- In retrospect, my "Certainty of the Bayesian Fortune-Teller" is a wordy glossy on part of this great paper.]]></description>
<dc:subject>to:NB decision_theory probability bayesianism ellsberg.daniel rationality</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e7e72458dccd/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:ellsberg.daniel"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:rationality"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.mitpressjournals.org/doi/abs/10.1162/NECO_a_00912#.WH0fI7GZOL8">
    <title>Active Inference: A Process Theory | Neural Computation | MIT Press Journals</title>
    <dc:date>2017-01-16T19:49:58+00:00</dc:date>
    <link>http://www.mitpressjournals.org/doi/abs/10.1162/NECO_a_00912#.WH0fI7GZOL8</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This article describes a process theory based on active inference and belief propagation. Starting from the premise that all neuronal processing (and action selection) can be explained by maximizing Bayesian model evidence—or minimizing variational free energy—we ask whether neuronal responses can be described as a gradient descent on variational free energy. Using a standard (Markov decision process) generative model, we derive the neuronal dynamics implicit in this description and reproduce a remarkable range of well-characterized neuronal phenomena. These include repetition suppression, mismatch negativity, violation responses, place-cell activity, phase precession, theta sequences, theta-gamma coupling, evidence accumulation, race-to-bound dynamics, and transfer of dopamine responses. Furthermore, the (approximately Bayes’ optimal) behavior prescribed by these dynamics has a degree of face validity, providing a formal explanation for reward seeking, context learning, and epistemic foraging. Technically, the fact that a gradient descent appears to be a valid description of neuronal activity means that variational free energy is a Lyapunov function for neuronal dynamics, which therefore conform to Hamilton’s principle of least action."

--- To be shot after a fair trial.]]></description>
<dc:subject>to:NB neural_coding_and_decoding neuroscience bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:d067a1075404/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neural_coding_and_decoding"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neuroscience"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1007/s11229-015-0800-7">
    <title>A dilemma for the imprecise bayesian - Springer</title>
    <dc:date>2016-05-14T18:05:28+00:00</dc:date>
    <link>http://link.springer.com/article/10.1007/s11229-015-0800-7</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Many philosophers regard the imprecise credence framework as a more realistic model of probabilistic inferences with imperfect empirical information than the traditional precise credence framework. Hence, it is surprising that the literature lacks any discussion on how to update one’s imprecise credences when the given evidence itself is imprecise. To fill this gap, I consider two updating principles. Unfortunately, each of them faces a serious problem. The first updating principle, which I call “generalized conditionalization,” sometimes forces an agent to change her imprecise degrees of belief even though she does not have new evidence. The second updating principle, which I call “the generalized dynamic Keynesian model,” may result in a very precise credal state although the agent does not have sufficiently strong evidence to justify such an informative doxastic state. This means that it is much more difficult to come up with an acceptable updating principle for the imprecise credence framework than one might have thought it would be."]]></description>
<dc:subject>to:NB statistics bayesianism epistemology probability philosophy_of_science</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1214f6165de7/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:epistemology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1507.01059">
    <title>[1507.01059] Remarks on kernel Bayes' rule</title>
    <dc:date>2015-08-06T15:32:03+00:00</dc:date>
    <link>http://arxiv.org/abs/1507.01059</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Kernel Bayes' rule has been proposed as a nonparametric kernel-based method to realize Bayesian inference in reproducing kernel Hilbert spaces. However, we demonstrate both theoretically and experimentally that the prediction result by kernel Bayes' rule is in some cases unnatural. We consider that this phenomenon is in part due to the fact that the assumptions in kernel Bayes' rule do not hold in general."

--- The point about the "kernel posterior" being insensitive to the prior seems the strongest to me; so strong that despite reading the proof I am not sure it's right, as opposed to an algebraic mistake I'm missing.]]></description>
<dc:subject>to:NB bayesianism kernel_methods hilbert_space computational_statistics statistics have_read</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:6f37180f07be/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kernel_methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hilbert_space"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://rstl.royalsocietypublishing.org/content/53/370.short">
    <title>An Essay towards Solving a Problem in the Doctrine of Chances. By the Late Rev. Mr. Bayes, F. R. S. Communicated by Mr. Price, in a Letter to John Canton, A. M. F. R. S.</title>
    <dc:date>2015-06-22T18:31:23+00:00</dc:date>
    <link>http://rstl.royalsocietypublishing.org/content/53/370.short</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[Sometimes, reading the classics is so very, very _not_ worth the trouble.  The things I do for these referee reports...]]></description>
<dc:subject>have_read probability statistics foundations_of_statistics bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c4756490e987/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.pnas.org/content/112/12/3788.abstract.html?etoc">
    <title>Latent structure in random sequences drives neural learning toward a rational bias</title>
    <dc:date>2015-03-28T14:12:16+00:00</dc:date>
    <link>http://www.pnas.org/content/112/12/3788.abstract.html?etoc</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["People generally fail to produce random sequences by overusing alternating patterns and avoiding repeating ones—the gambler’s fallacy bias. We can explain the neural basis of this bias in terms of a biologically motivated neural model that learns from errors in predicting what will happen next. Through mere exposure to random sequences over time, the model naturally develops a representation that is biased toward alternation, because of its sensitivity to some surprisingly rich statistical structure that emerges in these random sequences. Furthermore, the model directly produces the best-fitting bias-gain parameter for an existing Bayesian model, by which we obtain an accurate fit to the human data in random sequence production. These results show that our seemingly irrational, biased view of randomness can be understood instead as the perfectly reasonable response of an effective learning mechanism to subtle statistical structure embedded in random sequences."

--- I'll be very surprised to see how "overusing alternating patterns and avoiding repeating ones" can be made to fall out from any sort of Bayesian model that doesn't have it built in from the start.  (In particular, the natural Bayesian model is that an unknown sequence is exchangeable, which would tend to imply more repeating patterns.)]]></description>
<dc:subject>to:NB probability cognitive_science bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1a1ecd1c649e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://pss.sagepub.com/content/24/12/2351">
    <title>How Robust Are Probabilistic Models of Higher-Level Cognition?</title>
    <dc:date>2015-02-28T05:00:16+00:00</dc:date>
    <link>http://pss.sagepub.com/content/24/12/2351</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["An increasingly popular theory holds that the mind should be viewed as a near-optimal or rational engine of probabilistic inference, in domains as diverse as word learning, pragmatics, naive physics, and predictions of the future. We argue that this view, often identified with Bayesian models of inference, is markedly less promising than widely believed, and is undermined by post hoc practices that merit wholesale reevaluation. We also show that the common equation between probabilistic and rational or optimal is not justified."]]></description>
<dc:subject>psychology cognitive_science bayesianism marcus.gary_f. have_read in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:904ad4d04837/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:marcus.gary_f."/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1411.3984">
    <title>[1411.3984] Qualitative Robustness in Bayesian Inference</title>
    <dc:date>2015-01-23T13:44:31+00:00</dc:date>
    <link>http://arxiv.org/abs/1411.3984</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We develop a framework for quantifying the sensitivity of the distribution of posterior distributions with respect to perturbations of the prior and data generating distributions in the limit when the number of data points grows towards infinity. In this generalization of Hampel and Cuevas' notion of qualitative robustness to Bayesian inference, posterior distributions are analyzed as measure-valued random variables (measures randomized through the data) and their robustness is quantified using the total variation, Prokhorov, and Ky Fan metrics. Our results show that (1) the assumption that the prior has Kullback-Leibler support at the parameter value generating the data, classically used to prove consistency, can also be used to prove the non-robustness of posterior distributions with respect to infinitesimal perturbations (in total variation metric) of the class of priors satisfying that assumption, (2) for a prior which has global Kullback-Leibler support on a space which is not totally bounded, we can establish non qualitative robustness and (3) consistency and robustness are, to some degree, antagonistic requirements and a careful selection of the prior is important if both properties (or their approximations) are to be achieved. 
"The mechanisms supporting our results are different and complementary to those discovered by Hampel and developed by Cuevas, and also indicate that misspecification generates non qualitative robustness."]]></description>
<dc:subject>bayesianism bayesian_consistency misspecification statistics in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:4cb2f5132afc/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1412.3442">
    <title>[1412.3442] Posterior predictive p-values and the convex order</title>
    <dc:date>2015-01-20T13:23:47+00:00</dc:date>
    <link>http://arxiv.org/abs/1412.3442</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Posterior predictive p-values are a common approach to Bayesian model-checking. This article analyses their frequency behaviour, that is, their distribution when the parameters and the data are drawn from the prior and the model respectively. We show that the family of possible distributions is exactly described as the distributions that are less variable than uniform on [0,1], in the convex order. In general, p-values with such a property are not conservative, and we illustrate how the theoretical worst-case error rate for false rejection can occur in practice. We describe how to correct the p-values to recover conservatism in several common scenarios, for example, when interpreting a single p-value or when combining multiple p-values into an overall score of significance. We also handle the case where the p-value is estimated from posterior samples obtained from techniques such as Markov Chain or Sequential Monte Carlo. Our results place posterior predictive p-values in a much clearer theoretical framework, allowing them to be used with more assurance."]]></description>
<dc:subject>to:NB to_read model_checking bayesianism re:phil-of-bayes_paper hypothesis_testing p-values statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:de3ce3593c11/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:model_checking"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hypothesis_testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:p-values"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1410.7600">
    <title>[1410.7600] Discussion of: `Frequentist coverage of adaptive nonparametric Bayesian credible sets'</title>
    <dc:date>2015-01-20T02:55:54+00:00</dc:date>
    <link>http://arxiv.org/abs/1410.7600</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>confidence_sets bayesianism have_read bayesian_consistency nonparametrics statistics in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:b0b4af48490f/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:confidence_sets"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:nonparametrics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1501.03326">
    <title>[1501.03326] Unbiased Bayes for Big Data: Paths of Partial Posteriors</title>
    <dc:date>2015-01-19T15:18:25+00:00</dc:date>
    <link>http://arxiv.org/abs/1501.03326</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Bayesian inference proceeds based on expectations of certain functions with respect to the posterior. Markov Chain Monte Carlo is a fundamental tool to compute these expectations. However, its feasibility is being challenged in the era of so called Big Data as all data needs to be processed in every iteration. Realising that such simulation is an unnecessarily hard problem if the goal is estimation, we construct a computationally scalable methodology that allows unbiased estimation of the required expectations -- without explicit simulation from the full posterior. The average computational complexity of our scheme is sub-linear in the size of the dataset and its variance is straightforward to control, leading to algorithms that are provably unbiased and naturally arrive at a desired error tolerance. We demonstrate the utility and generality of the methodology on a range of common statistical models applied to large scale benchmark datasets."]]></description>
<dc:subject>to:NB monte_carlo bayesianism computational_statistics statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a9654403ca29/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:monte_carlo"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1412.3730">
    <title>[1412.3730] Inconsistency of Bayesian Inference for Misspecified Linear Models, and a Proposal for Repairing It</title>
    <dc:date>2014-12-17T18:05:59+00:00</dc:date>
    <link>http://arxiv.org/abs/1412.3730</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We empirically show that Bayesian inference can be inconsistent under misspecification in simple linear regression problems, both in a model averaging/selection and in a Bayesian ridge regression setting. We use the standard linear model, which assumes homoskedasticity, whereas the data are heteroskedastic, and observe that the posterior puts its mass on ever more high-dimensional models as the sample size increases. To remedy the problem, we equip the likelihood in Bayes' theorem with an exponent called the learning rate, and we propose the Safe Bayesian method to learn the learning rate from the data. SafeBayes tends to select small learning rates as soon the standard posterior is not `cumulatively concentrated', and its results on our data are quite encouraging."]]></description>
<dc:subject>to_read linear_regression bayesianism bayesian_consistency misspecification statistics grunwald.peter in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:9de43c988a99/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:linear_regression"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:grunwald.peter"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://philsci-archive.pitt.edu/3165/">
    <title>Ignorance and Indifference - PhilSci-Archive</title>
    <dc:date>2014-09-03T18:47:29+00:00</dc:date>
    <link>http://philsci-archive.pitt.edu/3165/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The epistemic state of complete ignorance is not a probability distribution. In it, we assign the same, unique ignorance degree of belief to any contingent outcome and each of its contingent, disjunctive parts. That this is the appropriate way to represent complete ignorance is established by two instruments, each individually strong enough to identify this state. They are the principle of indifference (“PI”) and the notion that ignorance is invariant under certain redescriptions of the outcome space, here developed into the “principle of invariance of ignorance” (“PII”). Both instruments are so innocuous as almost to be platitudes. Yet the literature in probabilistic epistemology has misdiagnosed them as paradoxical or defective since they generate inconsistencies when conjoined with the assumption that an epistemic state must be a probability distribution. To underscore the need to drop this assumption, I express PII in its most defensible form as relating symmetric descriptions and show that paradoxes still arise if we assume the ignorance state to be a probability distribution. By separating out the different properties that characterize a probability measure, I show that the ignorance state is incompatible with each of the additivity and the dynamics of Bayesian conditionalization of the probability calculus."]]></description>
<dc:subject>to:NB philosophy_of_science epistemology bayesianism foundations_of_statistics norton.john_d.</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:452062adf5f7/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:epistemology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:norton.john_d."/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1007/s11229-013-0286-0">
    <title>The main two arguments for probabilism are flawed - Springer</title>
    <dc:date>2014-06-15T15:19:26+00:00</dc:date>
    <link>http://link.springer.com/article/10.1007/s11229-013-0286-0</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Probabilism, the view that agents have numerical degrees of beliefs that conform to the axioms of probability, has been defended by the vast majority of its proponents by way of either of two arguments, the Dutch Book Argument and the Representation Theorems Argument. In this paper I argue that both arguments are flawed. The Dutch Book Argument is based on an unwarranted, ad hoc premise that cannot be dispensed with. The Representation Theorems Argument hinges on an invalid implication."

--- Ehh.  The argument against the Dutch Book Argument is that if the agent pays b for a bet that pays off $S or $0, _positing_ that their probability is p=b/S illegitimately smuggles in the additivity axiom.  Similarly for the representation theorems, even if there's a unique set of Kolmogorovian probabilities corresponding to preferences over lotteries, you could always come up with other, non-Kolmogorovian weights on states-of-the-world and other utilties to represent the same preferences.  These seem very weak to me (unlike "why do we care about these imaginary bets and lotteries?")]]></description>
<dc:subject>to:NB bayesianism decision_theory have_read my_initial_skeptical_coloration_became_on_examination_a_permanent_stain</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:19c99766b6c3/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:my_initial_skeptical_coloration_became_on_examination_a_permanent_stain"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2374040">
    <title>Posterior-Hacking: Selective Reporting Invalidates Bayesian Results Also by Uri Simonsohn :: SSRN</title>
    <dc:date>2014-03-26T23:38:16+00:00</dc:date>
    <link>http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2374040</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Many believe that Bayesian statistics are robust to p-hacking. Many are wrong. In this paper I show with simulations and actual data that the two Bayesian approaches that have been proposed within Psychology, Bayesian inference and Bayes factors, are as invalidated by selective reporting as p-values are. Going Bayesian may offer some benefits, providing a solution to selective reporting is not one of them. Required disclosure is the only solution."]]></description>
<dc:subject>to:NB statistics bad_data_analysis bayesianism hypothesis_testing meta-analysis re:neutral_model_of_inquiry have_read to:blog</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:8d1989526828/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bad_data_analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hypothesis_testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:meta-analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:neutral_model_of_inquiry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1403.4630">
    <title>[1403.4630] Penalising model component complexity: A principled, practical approach to constructing priors</title>
    <dc:date>2014-03-22T19:27:48+00:00</dc:date>
    <link>http://arxiv.org/abs/1403.4630</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The issue of setting prior distributions on model parameters, or to attribute uncertainty for model parameters, is a difficult issue in applied Bayesian statistics. Although the prior distribution should ideally encode the users' prior knowledge about the parameters, this level of knowledge transfer seems to be unattainable in practice and applied statisticians are forced to search for a "default" prior. Despite the development of objective priors, which are only available explicitly for a small number of highly restricted model classes, the applied statistician has few practical guidelines to follow when choosing the priors. An easy way out of this dilemma is to re-use prior choices of others, with an appropriate reference. 
"In this paper, we introduce a new concept for constructing prior distributions. We exploit the natural nested structure inherent to many model components, which defines the model component to be a flexible extension of a base model. Proper priors are defined to penalise the complexity induced by deviating from the simpler base model and are formulated after the input of a user-defined scaling parameter for that model component. These priors are invariant to reparameterisations, have a natural connection to Jeffreys' priors, are designed to support Occam's razor and seem to have excellent robustness properties, all which are highly desirable and allow us to use this approach to define default prior distributions."

- Christian Robert is enthusiastic, FWIW: http://xianblog.wordpress.com/2014/04/01/penalising-model-component-complexity/]]></description>
<dc:subject>to:NB bayesianism statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:374aa255c4b0/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1211.1530">
    <title>[1211.1530] Conditional inferential models: combining information for prior-free probabilistic inference</title>
    <dc:date>2014-02-18T00:35:27+00:00</dc:date>
    <link>http://arxiv.org/abs/1211.1530</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The inferential model (IM) framework provides valid prior-free probabilistic inference by focusing on predicting unobserved auxiliary variables. But, efficient IM-based inference can be challenging when the auxiliary variable is of higher dimension than the parameter. Here we show that features of the auxiliary variable are often fully observed and, in such cases, a simultaneous dimension reduction and information aggregation can be achieved by conditioning. This proposed conditioning strategy leads to efficient IM inference, and casts new light on Fisher's notions of sufficiency, conditioning, and also Bayesian inference. A differential equation-driven selection of a conditional association is developed, and validity of the conditional IM is proved under some conditions. For problems that do not admit a valid conditional IM of the standard form, we propose a more flexible class of conditional IMs based on localization. Examples of local conditional IMs in a bivariate normal model and a normal variance components model are also given."]]></description>
<dc:subject>to:NB statistics bayesianism foundations_of_statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:fd87f400f6a5/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.tandfonline.com/doi/abs/10.1198/jcgs.2011.09090#.Uu_-Dv0aAlM">
    <title>Taylor &amp; Francis Online :: Direct Sampling - Journal of Computational and Graphical Statistics - Volume 20, Issue 3</title>
    <dc:date>2014-02-03T20:39:16+00:00</dc:date>
    <link>http://www.tandfonline.com/doi/abs/10.1198/jcgs.2011.09090#.Uu_-Dv0aAlM</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In recent years, Markov chain Monte Carlo (MCMC) methods have been used to provide a full Bayesian analysis both when the posterior distribution of interest is analytically intractable, and it is not known how to draw independent samples. In this article, a non-MCMC approach to sampling from posterior distributions is developed and illustrated. Some sampling problems, now thought to be best handled by MCMC methods alone, are tackled efficiently via independent samples. This article has supplementary material online."

--- Hah, they don't see the trick...]]></description>
<dc:subject>to:NB have_skimmed not_quite_scooped monte_carlo bayesianism re:fitness_sampling</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:4fc69f37bc66/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_skimmed"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:not_quite_scooped"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:monte_carlo"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:fitness_sampling"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://jmlr.org/papers/v14/fukumizu13a.html">
    <title>Kernel Bayes' Rule: Bayesian Inference with Positive Definite Kernels</title>
    <dc:date>2014-01-28T01:20:00+00:00</dc:date>
    <link>http://jmlr.org/papers/v14/fukumizu13a.html</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A kernel method for realizing Bayes' rule is proposed, based on representations of probabilities in reproducing kernel Hilbert spaces. Probabilities are uniquely characterized by the mean of the canonical map to the RKHS. The prior and conditional probabilities are expressed in terms of RKHS functions of an empirical sample: no explicit parametric model is needed for these quantities. The posterior is likewise an RKHS mean of a weighted sample. The estimator for the expectation of a function of the posterior is derived, and rates of consistency are shown. Some representative applications of the kernel Bayes' rule are presented, including Bayesian computation without likelihood and filtering with a nonparametric state-space model."]]></description>
<dc:subject>to:NB statistics bayesianism kernel_methods hilbert_space</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f678a5bafb64/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kernel_methods"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hilbert_space"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1312.2974">
    <title>[1312.2974] A path-integral approach to Bayesian inference for inverse problems using the semiclassical approximation</title>
    <dc:date>2014-01-16T00:06:44+00:00</dc:date>
    <link>http://arxiv.org/abs/1312.2974</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We demonstrate how path integrals often used in problems of theoretical physics can be adapted to provide a machinery for performing Bayesian inference in function spaces. Such inference comes about naturally in the study of inverse problems of recovering continuous (infinite dimensional) coefficient functions from ordinary or partial differential equations (ODE, PDE), a problem which is typically ill-posed. Regularization of these problems using ℓ2 function spaces (Tikhonov regularization) is equivalent to Bayesian probabilistic inference, using a Gaussian prior. The Bayesian interpretation of inverse problem regularization is useful since it allows one to quantify and characterize error and degree of precision in the solution of inverse problems, as well as examine assumptions made in solving the problem -- namely whether the subjective choice of regularization is compatible with prior knowledge. Using path-integral formalism, Bayesian inference can be explored through various perturbative techniques, such as the semiclassical approximation, which we use in this manuscript. Perturbative path-integral approaches, while offering alternatives to computational approaches like Markov-Chain-Monte-Carlo (MCMC), also provide natural starting points for MCMC methods that can be used to refine perturbative approximations. In this manuscript, we illustrate a path-integral formulation for inverse problems and demonstrate it on an inverse problem in membrane biophysics as well as inverse problems in potential theory involving the Poisson equation."]]></description>
<dc:subject>inverse_problems bayesianism statistics monte_carlo kith_and_kin savage.van feynman_diagrams_and_path_integrals in_NB path_integrals_for_classical_stochastic_processes</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:4b50ca727672/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:inverse_problems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:monte_carlo"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kith_and_kin"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:savage.van"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:feynman_diagrams_and_path_integrals"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:path_integrals_for_classical_stochastic_processes"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1311.0072">
    <title>[1311.0072] Bayesian inference as iterated random functions with applications to sequential inference in graphical models</title>
    <dc:date>2013-12-17T18:25:17+00:00</dc:date>
    <link>http://arxiv.org/abs/1311.0072</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We propose a general formalism of iterated random functions with semigroup property, under which exact and approximate Bayesian posterior updates can be viewed as specific instances. A convergence theory for iterated random functions is presented. As an application of the general theory we analyze convergence behaviors of exact and approximate message-passing algorithms that arise in a sequential change point detection problem formulated via a latent variable directed graphical model. The sequential inference algorithm and its supporting theory are illustrated by simulated examples."]]></description>
<dc:subject>dynamical_systems bayesianism bayesian_consistency change-point_problem statistics stochastic_processes in_NB to_read</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:fa688060f5fb/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:dynamical_systems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:change-point_problem"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:stochastic_processes"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1312.0302">
    <title>[1312.0302] On the Equivalence between Bayesian and Classical Hypothesis Testing</title>
    <dc:date>2013-12-16T15:04:48+00:00</dc:date>
    <link>http://arxiv.org/abs/1312.0302</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["For hypotheses of the type H_0:theta=theta_0 vs H_1:theta ne theta_0 we demonstrate the equivalence of a Bayesian hypothesis test using a Bayes factor and the corresponding classical test, for a large class of models, which are detailed in the paper. In particular, we show that the role of the prior and critical region for the Bayes factor test is only to specify the type I error. This is their only role since, as we show, the power function of the Bayes factor test coincides exactly with that of the classical test, once the type I error has been fixed. 
"For more complex tests involving nuisance parameters, we recover the classical test by using Jeffreys prior on the nuisance parameters, while the prior on the hypothesized parameters can be arbitrary up to a large class. On the other hand, we show that using proper priors on the nuisance parameters results in a test with uniformly lower power than the classical test."

- Ouch.]]></description>
<dc:subject>have_read hypothesis_testing bayesianism statistics re:phil-of-bayes_paper in_NB to:blog</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1a00bd4bbb59/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hypothesis_testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:phil-of-bayes_paper"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.aeaweb.org/articles.php?doi=10.1257/aer.103.7.2790">
    <title>AER (103,7) p. 2790 - &amp;quot;Reverse Bayesianism&amp;quot;: A Choice-Based Theory of Growing Awareness</title>
    <dc:date>2013-12-04T20:43:31+00:00</dc:date>
    <link>http://www.aeaweb.org/articles.php?doi=10.1257/aer.103.7.2790</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This article introduces a new approach to modeling the expanding universe of decision makers in the wake of growing awareness, and invokes the axiomatic approach to model the evolution of decision makers' beliefs as awareness grows. The expanding universe is accompanied by extension of the set of acts, the preference relations over which are linked by a new axiom, invariant risk preferences, asserting that the ranking of lotteries is independent of the set of acts under consideration. The main results are representation theorems and rules for updating beliefs over expanding state spaces and events that have the flavor of "reverse Bayesianism."]]></description>
<dc:subject>to:NB decision_theory bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c1f28b9bac38/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.jstor.org/discover/10.1086/673249">
    <title>Bayesian Orgulity [JSTOR: Philosophy of Science, Vol. 80, No. 4 (October 2013), pp. 483-503]</title>
    <dc:date>2013-10-11T22:16:45+00:00</dc:date>
    <link>http://www.jstor.org/discover/10.1086/673249</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A piece of folklore enjoys some currency among philosophical Bayesians, according to which Bayesian agents who, intuitively speaking, spread their credence over the entire space of available hypotheses are certain to converge to the truth. The goals of the current discussion are to show that that kernel of truth in this folklore is in some ways fairly small and to argue that Bayesian convergence-to-the-truth results are a liability for Bayesianism as an account of rationality since they render a certain sort of arrogance rationally mandatory."]]></description>
<dc:subject>bayesianism bayesian_consistency philosophy_of_science statistics have_read in_NB to:blog</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:dd395074fc55/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www-cs.stanford.edu/people/slingamn/philosophy/frequentism_positivism/frequentism_positivism.pdf">
    <title>Frequentism as a Positivism</title>
    <dc:date>2013-08-09T20:50:37+00:00</dc:date>
    <link>https://www-cs.stanford.edu/people/slingamn/philosophy/frequentism_positivism/frequentism_positivism.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["I explore an alternate clarification of the idea of frequency probabil- ity, called frequency judgment. I then distinguish three distinct senses of probability — physical chance, frequency judgment, and subjective credence — and propose that they have a hierarchical relationship. Finally, I claim that this three-tiered view can dissolve various para- doxes associated with the interpretation of probability."]]></description>
<dc:subject>to:NB have_skimmed foundations_of_probability philosophy_of_science bayesianism</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:099ea2610c94/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_skimmed"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:philosophy_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.jneurosci.org/content/33/26/10887.short?rss=1">
    <title>The Human Brain Encodes Event Frequencies While Forming Subjective Beliefs</title>
    <dc:date>2013-07-17T17:54:57+00:00</dc:date>
    <link>http://www.jneurosci.org/content/33/26/10887.short?rss=1</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["To make adaptive choices, humans need to estimate the probability of future events. Based on a Bayesian approach, it is assumed that probabilities are inferred by combining a priori, potentially subjective, knowledge with factual observations, but the precise neurobiological mechanism remains unknown. Here, we study whether neural encoding centers on subjective posterior probabilities, and data merely lead to updates of posteriors, or whether objective data are encoded separately alongside subjective knowledge. During fMRI, young adults acquired prior knowledge regarding uncertain events, repeatedly observed evidence in the form of stimuli, and estimated event probabilities. Participants combined prior knowledge with factual evidence using Bayesian principles. Expected reward inferred from prior knowledge was encoded in striatum. BOLD response in specific nodes of the default mode network (angular gyri, posterior cingulate, and medial prefrontal cortex) encoded the actual frequency of stimuli, unaffected by prior knowledge. In this network, activity increased with frequencies and thus reflected the accumulation of evidence. In contrast, Bayesian posterior probabilities, computed from prior knowledge and stimulus frequencies, were encoded in bilateral inferior frontal gyrus. Here activity increased for improbable events and thus signaled the violation of Bayesian predictions. Thus, subjective beliefs and stimulus frequencies were encoded in separate cortical regions. The advantage of such a separation is that objective evidence can be recombined with newly acquired knowledge when a reinterpretation of the evidence is called for. Overall this study reveals the coexistence in the brain of an experience-based system of inference and a knowledge-based system of inference."

--- Errr, what's specifically Bayesian here?]]></description>
<dc:subject>to:NB bayesianism experimental_psychology neuroscience fmri</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f084eb4799ee/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:experimental_psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:neuroscience"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:fmri"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1306.6430">
    <title>[1306.6430] A General Framework for Updating Belief Distributions</title>
    <dc:date>2013-06-30T03:40:59+00:00</dc:date>
    <link>http://arxiv.org/abs/1306.6430</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We propose a general framework for Bayesian inference that does not require the specification of a complete probability model, or likelihood, for the data. As data sets become larger and systems under investigation more complex it is increasingly challenging for Bayesian analysts to attempt to model the true data generating mechanism. Moreover, when the object of interest is a low dimensional statistic, such as a mean or median, it is cumbersome to have to achieve this via a complete model for the whole data distribution. If Bayesian analysis is to keep pace with modern applications it will need to forsake the notion that it is either possible or desirable to model the complete data distribution. Our proposed framework uses loss-functions to connect information in the data to statistics of interest. The updating of beliefs then follows from a decision theoretic approach involving cumulative loss functions. Importantly, the procedure coincides with Bayesian updating when a true likelihood is given, yet provides coherent subjective inference in much more general settings. We demonstrate our approach in important application areas for which Bayesian inference is problematic including variable selection in survival analysis models and inference on a set of quantiles of a sampling distribution. Connections to other inference frameworks are highlighted."]]></description>
<dc:subject>high-dimensional_statistics bayesianism statistics estimation have_skimmed bayesian_consistency in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:840cce4c1bb8/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:high-dimensional_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:estimation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_skimmed"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://ba.stat.cmu.edu/abstracts/Scutari.php">
    <title>On the Prior and Posterior Distributions Used in Graphical Modelling</title>
    <dc:date>2013-06-27T15:24:13+00:00</dc:date>
    <link>http://ba.stat.cmu.edu/abstracts/Scutari.php</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Graphical model learning and inference are often performed using Bayesian techniques. In particular, learning is usually performed in two separate steps. First, the graph structure is learned from the data; then the parameters of the model are estimated conditional on that graph structure. While the probability distributions involved in this second step have been studied in depth, the ones used in the first step have not been explored in as much detail. In this paper, we will study the prior and posterior distributions defined over the space of the graph structures for the purpose of learning the structure of a graphical model. In particular, we will provide a characterisation of the behaviour of those distributions as a function of the possible edges of the graph. We will then use the properties resulting from this characterisation to define measures of structural variability for both Bayesian and Markov networks, and we will point out some of their possible applications."]]></description>
<dc:subject>to:NB bayesianism graphical_models statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:7877db817a1e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:graphical_models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1306.3092">
    <title>[1306.3092] Marginal inferential models: prior-free probabilistic inference on interest parameters</title>
    <dc:date>2013-06-27T15:00:10+00:00</dc:date>
    <link>http://arxiv.org/abs/1306.3092</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Inferential models (IMs) provide a general framework for prior-free, frequency-calibrated, posterior probabilistic inference. The fundamental idea is to use auxiliary variables to reason with uncertainty about the parameter of interest. When nuisance parameters are present, a marginalization step can reduce the dimension of the auxiliary variable, which in turn leads to more efficient inference. For regular problems, exact and efficient marginalization can be achieved, and we prove that the marginal IM is valid. We show that our approach provides efficient marginal inference in several challenging problems, including a many-normal-means problem, and does not suffer from common marginalization paradoxes. In non-regular problems, we propose a generalized marginalization technique which is valid and also paradox-free. Details are given for two benchmark examples, namely, the Behrens--Fisher and gamma mean problems."]]></description>
<dc:subject>to:NB bayesianism statistics estimation foundations_of_statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:6e08670caa7f/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bayesianism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:estimation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_statistics"/>
</rdf:Bag></taxo:topics>
</item>
</rdf:RDF>