<?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.aeaweb.org/articles?id=10.1257/aer.20240246"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2503.11709"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2408.02841"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2310.17651#"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2310.05921"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2103.01802"/>
	<rdf:li rdf:resource="http://frankackerman.com/worst-case-economics/"/>
	<rdf:li rdf:resource="https://www.jmlr.org/papers/v23/20-1165.html"/>
	<rdf:li rdf:resource="https://mkcamara.github.io/ctc.pdf"/>
	<rdf:li rdf:resource="https://www.aeaweb.org/articles?id=10.1257/aer.20180517"/>
	<rdf:li rdf:resource="https://doi.org/10.1257/jep.35.3.157"/>
	<rdf:li rdf:resource="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3444678"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2104.10573"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/2103.01115"/>
	<rdf:li rdf:resource="https://www.pnas.org/content/118/15/e2022886118"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1911.05873"/>
	<rdf:li rdf:resource="https://www.aeaweb.org/articles?id=10.1257/aer.20190390"/>
	<rdf:li rdf:resource="https://www.cambridge.org/core/journals/evolutionary-human-sciences/article/fitnessmaximizers-employ-pessimistic-probability-weighting-for-decisions-under-risk/FCF743180A566332C8AF9F7E7406AB43"/>
	<rdf:li rdf:resource="https://wwnorton.com/books/9781324004776/about-the-book/description"/>
	<rdf:li rdf:resource="https://www.cambridge.org/us/academic/subjects/economics/econometrics-statistics-and-mathematical-economics/ignorance-and-uncertainty?format=PB"/>
	<rdf:li rdf:resource="https://www.nature.com/articles/s41567-019-0732-0"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1909.06853"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1906.07801"/>
	<rdf:li rdf:resource="https://arxiv.org/abs/1704.04118"/>
	<rdf:li rdf:resource="http://danielcsaba.com/content/bayes_miss_Oct2018.pdf"/>
	<rdf:li rdf:resource="http://www.pnas.org/content/115/44/E10387?etoc="/>
	<rdf:li rdf:resource="https://www.jacobinmag.com/2015/09/vietnam-war-cambodia-ellsberg-pentagon-papers-kissinger"/>
	<rdf:li rdf:resource="https://www.aeaweb.org/articles?id=10.1257/aer.20161079"/>
	<rdf:li rdf:resource="https://link.springer.com/article/10.1007/s11238-017-9600-5"/>
	<rdf:li rdf:resource="http://www.pnas.org/content/114/32/8499.abstract.html"/>
	<rdf:li rdf:resource="https://link.springer.com/article/10.1007/s11238-017-9592-1?wt_mc=alerts.TOCjournals"/>
	<rdf:li rdf:resource="http://www.press.uchicago.edu/ucp/books/book/chicago/R/bo3639393.html"/>
	<rdf:li rdf:resource="http://www.jstor.org/stable/1884324?seq=1#page_scan_tab_contents"/>
	<rdf:li rdf:resource="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2729537"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11229-015-0846-6?wt_mc=alerts.TOCjournals"/>
	<rdf:li rdf:resource="https://www.aeaweb.org/articles?id=10.1257/aer.20150678"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11238-015-9526-8"/>
	<rdf:li rdf:resource="http://nostalgebraist.tumblr.com/post/143718406034/the-future-of-humanity-institute-seems-very"/>
	<rdf:li rdf:resource="https://orgtheory.wordpress.com/2015/06/26/picking-the-right-metric-from-college-ratings-to-the-cold-war/#more-29713"/>
	<rdf:li rdf:resource="http://laeuferpaar.de/Papers/ExpertsGroupDecisions.pdf"/>
	<rdf:li rdf:resource="http://mitpress.mit.edu/books/rational-action"/>
	<rdf:li rdf:resource="https://www.aeaweb.org/articles.php?doi=10.1257/jep.28.4.213"/>
	<rdf:li rdf:resource="http://cowles.econ.yale.edu/P/cp/p10b/p1053.pdf"/>
	<rdf:li rdf:resource="http://statweb.stanford.edu/~cgates/PERSI/papers/thinking.pdf"/>
	<rdf:li rdf:resource="http://www.slate.com/articles/technology/bitwise/2014/07/roko_s_basilisk_the_most_terrifying_thought_experiment_of_all_time.single.html"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1406.7563"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11238-013-9393-0"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11229-013-0286-0"/>
	<rdf:li rdf:resource="http://journal.sjdm.org/14/14206/jdm14206.html"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1402.6118"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1007/s11229-014-0398-1"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1402.4884"/>
	<rdf:li rdf:resource="http://global.oup.com/academic/product/risk-and-rationality-9780199672165"/>
	<rdf:li rdf:resource="http://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-022513-115553"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1312.1277"/>
	<rdf:li rdf:resource="http://www.aeaweb.org/articles.php?doi=10.1257/aer.103.7.2790"/>
	<rdf:li rdf:resource="http://asset2013.org/archivo/ficheros/Microeconomics/19-CUHADAROGLU-TUGCE.pdf"/>
	<rdf:li rdf:resource="http://www.pnas.org/content/110/39/E3704.abstract"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1307.0473"/>
	<rdf:li rdf:resource="http://www.aeaweb.org/articles.php?doi=10.1257/jel.51.2.496"/>
	<rdf:li rdf:resource="http://link.springer.com/article/10.1023%2FA%3A1022899518027"/>
	<rdf:li rdf:resource="http://noahpinionblog.blogspot.com/2013/05/bets-do-not-necessarily-reveal-beliefs.html"/>
	<rdf:li rdf:resource="http://www.mcsweeneys.net/articles/its-not-you-its-quantitative-cost-benefit-analysis#.URO1r9JCDZA.twitter"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1301.3840"/>
	<rdf:li rdf:resource="http://www.aeaweb.org/articles.php?doi=10.1257/jep.26.4.43"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1208.3213"/>
	<rdf:li rdf:resource="http://www-stat.stanford.edu/~cgates/PERSI/papers/thinking.pdf"/>
	<rdf:li rdf:resource="http://www.aeaweb.org/articles.php?doi=10.1257/jel.50.2.368"/>
	<rdf:li rdf:resource="http://infostructuralist.wordpress.com/2012/06/01/information-theory-in-economics-part-i-rational-inattention/"/>
	<rdf:li rdf:resource="http://arxiv.org/abs/1202.3079"/>
      </rdf:Seq>
    </items>
  </channel><item rdf:about="https://www.aeaweb.org/articles?id=10.1257/aer.20240246">
    <title>Robust Misspecified Models - American Economic Association</title>
    <dc:date>2026-04-09T13:11:47+00:00</dc:date>
    <link>https://www.aeaweb.org/articles?id=10.1257/aer.20240246</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This paper studies which misspecified models are likely to persist when decision-makers compare them with competing models. The main result characterizes such models based on two features that can be derived from primitives: The model's asymptotic accuracy in predicting the equilibrium distribution of observed outcomes and the "tightness" of the prior around such equilibria. Misspecified models can be robust, persisting against any arbitrary competing model—including the true model—despite decision-makers observing an infinite amount of data. Moreover, simple misspecified models equipped with entrenched priors can be more robust than complex correctly specified models."]]></description>
<dc:subject>decision_theory misspecification re:bayes_as_evol statistics in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f59185400440/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:bayes_as_evol"/>
	<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="https://arxiv.org/abs/2503.11709">
    <title>[2503.11709] Conformal Prediction and Human Decision Making</title>
    <dc:date>2025-04-14T15:38:24+00:00</dc:date>
    <link>https://arxiv.org/abs/2503.11709</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Methods to quantify uncertainty in predictions from arbitrary models are in demand in high-stakes domains like medicine and finance. Conformal prediction has emerged as a popular method for producing a set of predictions with specified average coverage, in place of a single prediction and confidence value. However, the value of conformal prediction sets to assist human decisions remains elusive due to the murky relationship between coverage guarantees and decision makers' goals and strategies. How should we think about conformal prediction sets as a form of decision support? We outline a decision theoretic framework for evaluating predictive uncertainty as informative signals, then contrast what can be said within this framework about idealized use of calibrated probabilities versus conformal prediction sets. Informed by prior empirical results and theories of human decisions under uncertainty, we formalize a set of possible strategies by which a decision maker might use a prediction set. We identify ways in which conformal prediction sets and posthoc predictive uncertainty quantification more broadly are in tension with common goals and needs in human-AI decision making. We give recommendations for future research in predictive uncertainty quantification to support human decision makers."]]></description>
<dc:subject>to:NB conformal_prediction gelman.andrew kith_and_kin via:rvenkat decision_theory decision-making</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:55bcde83ad7f/</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:conformal_prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:gelman.andrew"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kith_and_kin"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:rvenkat"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision-making"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2408.02841">
    <title>[2408.02841] Evaluating Posterior Probabilities: Decision Theory, Proper Scoring Rules, and Calibration</title>
    <dc:date>2024-08-21T17:38:25+00:00</dc:date>
    <link>https://arxiv.org/abs/2408.02841</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Most machine learning classifiers are designed to output posterior probabilities for the classes given the input sample. These probabilities may be used to make the categorical decision on the class of the sample; provided as input to a downstream system; or provided to a human for interpretation. Evaluating the quality of the posteriors generated by these system is an essential problem which was addressed decades ago with the invention of proper scoring rules (PSRs). Unfortunately, much of the recent machine learning literature uses calibration metrics -- most commonly, the expected calibration error (ECE) -- as a proxy to assess posterior performance. The problem with this approach is that calibration metrics reflect only one aspect of the quality of the posteriors, ignoring the discrimination performance. For this reason, we argue that calibration metrics should play no role in the assessment of posterior quality. Expected PSRs should instead be used for this job, preferably normalized for ease of interpretation. In this work, we first give a brief review of PSRs from a practical perspective, motivating their definition using Bayes decision theory. We discuss why expected PSRs provide a principled measure of the quality of a system's posteriors and why calibration metrics are not the right tool for this job. We argue that calibration metrics, while not useful for performance assessment, may be used as diagnostic tools during system development. With this purpose in mind, we discuss a simple and practical calibration metric, called calibration loss, derived from a decomposition of expected PSRs. We compare this metric with the ECE and with the expected score divergence calibration metric from the PSR literature and argue, using theoretical and empirical evidence, that calibration loss is superior to these two metrics."]]></description>
<dc:subject>calibration probability prediction decision_theory scoring_rules in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c76f79ade534/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:calibration"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:probability"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:scoring_rules"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2310.17651#">
    <title>[2310.17651] High-Dimensional Prediction for Sequential Decision Making</title>
    <dc:date>2023-12-10T02:33:04+00:00</dc:date>
    <link>https://arxiv.org/abs/2310.17651#</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We study the problem of making predictions of an adversarially chosen high-dimensional state that are unbiased subject to an arbitrary collection of conditioning events, with the goal of tailoring these events to downstream decision makers. We give efficient algorithms for solving this problem, as well as a number of applications that stem from choosing an appropriate set of conditioning events.
"For example, we can efficiently make predictions targeted at polynomially many decision makers, giving each of them optimal swap regret if they best-respond to our predictions. We generalize this to online combinatorial optimization, where the decision makers have a very large action space, to give the first algorithms offering polynomially many decision makers no regret on polynomially many subsequences that may depend on their actions and the context. We apply these results to get efficient no-subsequence-regret algorithms in extensive-form games (EFGs), yielding a new family of regret guarantees for EFGs that generalizes some existing EFG regret notions, e.g. regret to informed causal deviations, and is generally incomparable to other known such notions.
"Next, we develop a novel transparent alternative to conformal prediction for building valid online adversarial multiclass prediction sets. We produce class scores that downstream algorithms can use for producing valid-coverage prediction sets, as if these scores were the true conditional class probabilities. We show this implies strong conditional validity guarantees including set-size-conditional and multigroup-fair coverage for polynomially many downstream prediction sets. Moreover, our class scores can be guaranteed to have improved L2 loss, cross-entropy loss, and generally any Bregman loss, compared to any collection of benchmark models, yielding a high-dimensional real-valued version of omniprediction."]]></description>
<dc:subject>prediction decision_theory low-regret_learning roth.aaron conformal_prediction in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1d91278c068d/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:low-regret_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:roth.aaron"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:conformal_prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2310.05921">
    <title>[2310.05921] Conformal Decision Theory: Safe Autonomous Decisions from Imperfect Predictions</title>
    <dc:date>2023-11-09T19:19:41+00:00</dc:date>
    <link>https://arxiv.org/abs/2310.05921</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We introduce Conformal Decision Theory, a framework for producing safe autonomous decisions despite imperfect machine learning predictions. Examples of such decisions are ubiquitous, from robot planning algorithms that rely on pedestrian predictions, to calibrating autonomous manufacturing to exhibit high throughput and low error, to the choice of trusting a nominal policy versus switching to a safe backup policy at run-time. The decisions produced by our algorithms are safe in the sense that they come with provable statistical guarantees of having low risk without any assumptions on the world model whatsoever; the observations need not be I.I.D. and can even be adversarial. The theory extends results from conformal prediction to calibrate decisions directly, without requiring the construction of prediction sets. Experiments demonstrate the utility of our approach in robot motion planning around humans, automated stock trading, and robot manufacturing."]]></description>
<dc:subject>to_read conformal_prediction decision_theory jordan.michael_i. via:rvenkat in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:3f36ce96d5e8/</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:conformal_prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:jordan.michael_i."/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:rvenkat"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2103.01802">
    <title>[2103.01802] Median Optimal Treatment Regimes</title>
    <dc:date>2023-06-29T16:01:22+00:00</dc:date>
    <link>https://arxiv.org/abs/2103.01802</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Optimal treatment regimes are personalized policies for making a treatment decision based on subject characteristics, with the policy chosen to maximize some value. It is common to aim to maximize the mean outcome in the population, via a regime assigning treatment only to those whose mean outcome is higher under treatment versus control. However, the mean can be an unstable measure of centrality, resulting in imprecise statistical procedures, as well as unrobust decisions that can be overly influenced by a small fraction of subjects. In this work, we propose a new median optimal treatment regime that instead treats individuals whose conditional median is higher under treatment. This ensures that optimal decisions for individuals from the same group are not overly influenced either by (i) a small fraction of the group (unlike the mean criterion), or (ii) unrelated subjects from different groups (unlike marginal median/quantile criteria). We introduce a new measure of value, the Average Conditional Median Effect (ACME), which summarizes across-group median treatment outcomes of a policy, and which the median optimal treatment regime maximizes. After developing key motivating examples that distinguish median optimal treatment regimes from mean and marginal median optimal treatment regimes, we give a nonparametric efficiency bound for estimating the ACME of a policy, and propose a new doubly robust-style estimator that achieves the efficiency bound under weak conditions. To construct the median optimal treatment regime, we introduce a new doubly robust-style estimator for the conditional median treatment effect. Finite-sample properties are explored via numerical simulations and the proposed algorithm is illustrated using data from a randomized clinical trial in patients with HIV."]]></description>
<dc:subject>to_read decision_theory causal_inference kennedy.edward_h. kith_and_kin re:codename:one_law_for_the_lion_and_ox_is_oppression in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c25bcb7497ec/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:causal_inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kennedy.edward_h."/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kith_and_kin"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:codename:one_law_for_the_lion_and_ox_is_oppression"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://frankackerman.com/worst-case-economics/">
    <title>Worst-Case Economics: Extreme Events in Climate and Finance</title>
    <dc:date>2022-07-10T23:44:35+00:00</dc:date>
    <link>http://frankackerman.com/worst-case-economics/</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[--- Worth noting because of Bowles's endorsement.]]></description>
<dc:subject>in_NB books:noted economics decision_theory risk_vs_uncertainty books:suggest_to_library</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f8161043878a/</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:books:noted"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:books:suggest_to_library"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.jmlr.org/papers/v23/20-1165.html">
    <title>Approximate Information State for Approximate Planning and Reinforcement Learning in Partially Observed Systems</title>
    <dc:date>2022-03-25T18:56:36+00:00</dc:date>
    <link>https://www.jmlr.org/papers/v23/20-1165.html</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We propose a theoretical framework for approximate planning and learning in partially observed systems. Our framework is based on the fundamental notion of information state. We provide two definitions of information state---i) a function of history which is sufficient to compute the expected reward and predict its next value; ii) a function of the history which can be recursively updated and is sufficient to compute the expected reward and predict the next observation. An information state always leads to a dynamic programming decomposition. Our key result is to show that if a function of the history (called AIS) approximately satisfies the properties of the information state, then there is a corresponding approximate dynamic program. We show that the policy computed using this is approximately optimal with bounded loss of optimality. We show that several approximations in state, observation and action spaces in literature can be viewed as instances of AIS. In some of these cases, we obtain tighter bounds. A salient feature of AIS is that it can be learnt from data. We present AIS based multi-time scale policy gradient algorithms and detailed numerical experiments with low, moderate and high dimensional environments."

--- Hey, we're cited! :)
--- I really should've published that thesis chapter, but it's old enough to drink. :(]]></description>
<dc:subject>to_read prediction sufficiency decision_theory reinforcement_learning in_NB downloaded via:mraginsky</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:5485a68d723d/</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:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:sufficiency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:reinforcement_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:downloaded"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:mraginsky"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://mkcamara.github.io/ctc.pdf">
    <title>Computationally Tractable Choice</title>
    <dc:date>2022-02-08T14:31:56+00:00</dc:date>
    <link>https://mkcamara.github.io/ctc.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["I incorporate computational constraints into decision theory in order to capture how cognitive limitations affect behavior. I impose an axiom of computational tractability that only rules out behaviors that are thought to be fundamentally hard. I use this framework to better understand common behavioral heuristics: if choices are tractable and consistent with the expected utility axioms, then they are observationally equivalent to forms of choice bracketing. Then I show that a computationally-constrained decisionmaker can be objectively better off if she is willing to use heuristics that would not appear rational to an outside observer."

--- Simon and Lindblom vindicated.

--- ETA after reading: Yep, it does what it says.  If you like seeing reductions of SATISFIABILITY to optimization problems, here is a paper for you.
--- Also, the slippage of "rationality" in the last sentence of the abstract is kind of fascinating.  We _started_ by wanting to define "rational behavior" as being about effectively adapting means to ends; we had an intuition, inherited from 18th century philosophy, that calculating the expectation values in terms of rat orgasm equivalents would be a good way to adapt means to ends; we re-defined "rational behavior" as "acting as though one were calculating an expected number of rat orgasm equivalents"; now it turns out that that is provably an inferior way of adapting means to ends, and we have to worry about what it says about rationality.

--- Blog discussion: [http://bactra.org/weblog/1182.html#tractable]]]></description>
<dc:subject>decision_theory heuristics computational_complexity bounded_rationality via:suresh re:in_soviet_union_optimization_problem_solves_you have_read in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e9d5e55ce330/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:heuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:computational_complexity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bounded_rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:suresh"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:in_soviet_union_optimization_problem_solves_you"/>
	<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://www.aeaweb.org/articles?id=10.1257/aer.20180517">
    <title>The Risk-Adjusted Carbon Price - American Economic Association</title>
    <dc:date>2021-08-30T14:48:17+00:00</dc:date>
    <link>https://www.aeaweb.org/articles?id=10.1257/aer.20180517</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The social cost of carbon is the expected present value of damages from emitting one ton of carbon today. We use perturbation theory to derive an approximate tractable expression for this cost adjusted for climatic and economic risk. We allow for different aversion to risk and intertemporal fluctuations, skewness and dynamics in the risk distributions of climate sensitivity and the damage ratio, and correlated shocks. We identify prudence, insurance, and exposure effects, reproduce earlier analytical results, and offer analytical insights into numerical results on the effects of economic and damage ratio uncertainty and convex damages on the optimal carbon price."]]></description>
<dc:subject>to:NB decision_theory climate_change</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:5ec33f34551f/</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:climate_change"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://doi.org/10.1257/jep.35.3.157">
    <title>Statistical Significance, p-Values, and the Reporting of Uncertainty - American Economic Association</title>
    <dc:date>2021-08-05T02:56:36+00:00</dc:date>
    <link>https://doi.org/10.1257/jep.35.3.157</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The use of statistical significance and p-values has become a matter of substantial controversy in various fields using statistical methods. This has gone as far as some journals banning the use of indicators for statistical significance, or even any reports of p-values, and, in one case, any mention of confidence intervals. I discuss three of the issues that have led to these often-heated debates. First, I argue that in many cases, p-values and indicators of statistical significance do not answer the questions of primary interest. Such questions typically involve making (recommendations on) decisions under uncertainty. In that case, point estimates and measures of uncertainty in the form of confidence intervals or even better, Bayesian intervals, are often more informative summary statistics. In fact, in that case, the presence or absence of statistical significance is essentially irrelevant, and including them in the discussion may confuse the matter at hand. Second, I argue that there are also cases where testing null hypotheses is a natural goal and where p-values are reasonable and appropriate summary statistics. I conclude that banning them in general is counterproductive. Third, I discuss that the overemphasis in empirical work on statistical significance has led to abuse of p-values in the form of p-hacking and publication bias. The use of pre-analysis plans and replication studies, in combination with lowering the emphasis on statistical significance may help address these problems."]]></description>
<dc:subject>to:NB to_read statistics hypothesis_testing confidence_sets decision_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e1aad5699696/</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:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:hypothesis_testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:confidence_sets"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3444678">
    <title>Observational vs Experimental Data When Making Automated Decisions Using Machine Learning by Carlos Fernández-Loría, Foster Provost :: SSRN</title>
    <dc:date>2021-05-08T13:17:33+00:00</dc:date>
    <link>https://papers.ssrn.com/sol3/papers.cfm?abstract_id=3444678</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["With the recent explosion in both data and computing power, machine learning algorithms are increasingly used to make decisions automatically. These decisions are often causal in nature with the goal of improving an outcome by means of an intervention. Common examples include influencing someone's purchasing behavior with an advertisement or increasing customer retention with a special offer. Unfortunately, if these algorithms use observational data to estimate the effect of the interventions, the resulting estimates will likely suffer from confounding bias. Investing in experimental data offers a way to estimate effects without confounding bias, but such data are costly and may be in short supply. This paper addresses the question of whether it would be better to invest in costly experimental data or use the readily-available (but confounded) observational data. We present a theoretical comparison between the use of observational and experimental data when the goal is to build models to make automated intervention decisions. The key insight of the work is that optimizing to make the correct decision generally involves understanding whether a causal effect is above or below a given threshold, which is different from optimizing to reduce the magnitude of the bias in a causal-effect estimate. As a result, models trained with confounded observational data may lead to decisions that are just as good (or better) in certain scenarios, such as when larger causal effects are more likely to be overestimated or when the benefits of larger and cheaper data outweigh the detrimental effect of confounding. The theoretical results are tested by comparing the two approaches using the wide variety of benchmark data sets (7,700 in total) from the 2016 ACIC causal modeling competition. Finally, we suggest that sensitivity analysis may be used in practice to determine whether collecting experimental data to improve treatment assignments would be cost-effective, illustrating with a simple procedure that shows a ``Goldilocks effect'': in the illustration, the size of the experiment has to be just right for the investment to be worthwhile."]]></description>
<dc:subject>to:NB causal_inference experimental_design decision_theory via:arthegall</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:104863fd3ea6/</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:causal_inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:experimental_design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:arthegall"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/2104.10573">
    <title>[2104.10573] GEAR: On Optimal Decision Making with Auxiliary Data</title>
    <dc:date>2021-04-22T15:19:54+00:00</dc:date>
    <link>https://arxiv.org/abs/2104.10573</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Personalized optimal decision making, finding the optimal decision rule (ODR) based on individual characteristics, has attracted increasing attention recently in many fields, such as education, economics, and medicine. Current ODR methods usually require the primary outcome of interest in samples for assessing treatment effects, namely the experimental sample. However, in many studies, treatments may have a long-term effect, and as such the primary outcome of interest cannot be observed in the experimental sample due to the limited duration of experiments, which makes the estimation of ODR impossible. This paper is inspired to address this challenge by making use of an auxiliary sample to facilitate the estimation of ODR in the experimental sample. We propose an auGmented inverse propensity weighted Experimental and Auxiliary sample-based decision Rule (GEAR) by maximizing the augmented inverse propensity weighted value estimator over a class of decision rules using the experimental sample, with the primary outcome being imputed based on the auxiliary sample. The asymptotic properties of the proposed GEAR estimators and their associated value estimators are established. Simulation studies are conducted to demonstrate its empirical validity with a real AIDS application."

--- Last tag because the quality of the imputation seems both critical and nigh-impossible to check.]]></description>
<dc:subject>to:NB causal_inference missing_data decision_theory color_me_skeptical</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:6d219f0c104a/</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:causal_inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:missing_data"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<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/2103.01115">
    <title>[2103.01115] Structural models for policy-making: Coping with parametric uncertainty</title>
    <dc:date>2021-04-16T20:13:56+00:00</dc:date>
    <link>https://arxiv.org/abs/2103.01115</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The ex-ante evaluation of policies using structural econometric models is based on estimated parameters as a stand-in for the truth. This practice ignores uncertainty in the counterfactual policy predictions of the model. We develop a generic approach that deals with parametric uncertainty using uncertainty sets and frames model-informed policymaking as a decision problem under uncertainty. The seminal human capital investment model by Keane and Wolpin (1997) provides us with a well-known, influential, and empirically-grounded test case. We document considerable uncertainty in their policy predictions and highlight the resulting policy recommendations from using different formal rules on decision-making under uncertainty."]]></description>
<dc:subject>to:NB decision_theory risk_vs_uncertainty to_read</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:b721c477deff/</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:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.pnas.org/content/118/15/e2022886118">
    <title>Addressing partial identification in climate modeling and policy analysis | PNAS</title>
    <dc:date>2021-04-14T14:30:38+00:00</dc:date>
    <link>https://www.pnas.org/content/118/15/e2022886118</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Numerical simulations of the global climate system provide inputs to integrated assessment modeling for estimating the impacts of greenhouse gas mitigation and other policies to address global climate change. While essential tools for this purpose, computational climate models are subject to considerable uncertainty, including intermodel “structural” uncertainty. Structural uncertainty analysis has emphasized simple or weighted averaging of the outputs of multimodel ensembles, sometimes with subjective Bayesian assignment of probabilities across models. However, choosing appropriate weights is problematic. To use climate simulations in integrated assessment, we propose, instead, framing climate model uncertainty as a problem of partial identification, or “deep” uncertainty. This terminology refers to situations in which the underlying mechanisms, dynamics, or laws governing a system are not completely known and cannot be credibly modeled definitively even in the absence of data limitations in a statistical sense. We propose the min−max regret (MMR) decision criterion to account for deep climate uncertainty in integrated assessment without weighting climate model forecasts. We develop a theoretical framework for cost−benefit analysis of climate policy based on MMR, and apply it computationally with a simple integrated assessment model. We suggest avenues for further research."

--- On the one hand, Manski on partial identification is self-recommending.  On the other hand, it's a contributed paper in PNAS.

--- ETA after reading: Even Homer nods sometimes.  There's no actual need for, or use of, Manski's (pathbreaking) work on partial identification here.  Rather, the core of this is two quite simple, but sensible, recommendations: (i) if there are multiple good models, one way to represent uncertainty is simply to show the spread of different models, rather than trying to come up with weights that will force some sort of average; (ii) for each policy, find the optimal course of action, and then select the model-optimal policy which minimizes the maximum regret.  No connection to low-regret learning, which of course _does_ use weights.

--- It'd probably be good for the planet if we could leave Manski and Claire Monteleoni alone together on an island for a month.]]></description>
<dc:subject>climate_change decision_theory partial_identification modeling have_read to_blog in_NB manski.charles_f.</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:48bbbb0df653/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:climate_change"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:partial_identification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:modeling"/>
	<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:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:manski.charles_f."/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1911.05873">
    <title>[1911.05873] A Reduction from Reinforcement Learning to No-Regret Online Learning</title>
    <dc:date>2021-03-30T12:25:43+00:00</dc:date>
    <link>https://arxiv.org/abs/1911.05873</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We present a reduction from reinforcement learning (RL) to no-regret online learning based on the saddle-point formulation of RL, by which "any" online algorithm with sublinear regret can generate policies with provable performance guarantees. This new perspective decouples the RL problem into two parts: regret minimization and function approximation. The first part admits a standard online-learning analysis, and the second part can be quantified independently of the learning algorithm. Therefore, the proposed reduction can be used as a tool to systematically design new RL algorithms. We demonstrate this idea by devising a simple RL algorithm based on mirror descent and the generative-model oracle. For any γ-discounted tabular RL problem, with probability at least 1−δ, it learns an ϵ-optimal policy using at most Õ (||||log(1δ)(1−γ)4ϵ2) samples. Furthermore, this algorithm admits a direct extension to linearly parameterized function approximators for large-scale applications, with computation and sample complexities independent of ||,||, though at the cost of potential approximation bias."]]></description>
<dc:subject>to:NB reinforcement_learning low-regret_learning boots.byron gordon.geoff learning_theory decision_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e96a27d919c0/</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:reinforcement_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:low-regret_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:boots.byron"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:gordon.geoff"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:learning_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.aeaweb.org/articles?id=10.1257/aer.20190390">
    <title>A Theory of Chosen Preferences - American Economic Association</title>
    <dc:date>2021-01-28T17:28:45+00:00</dc:date>
    <link>https://www.aeaweb.org/articles?id=10.1257/aer.20190390</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We propose and develop a dynamic theory of endogenous preference formation in which people adopt worldviews that shape their judgments about their experiences. The framework highlights the role of mindset flexibility, a trait that determines the relative weights the decision-maker places on her current and anticipated worldviews when evaluating future outcomes. The theory generates rich behavioral dynamics, thereby illuminating a wide range of applications and providing potential explanations for a variety of observed phenomena."

--- This is a serious issue for both normative and positive theories of decision making.  (To give a small, personal example: when an ex told me she'd signed up for a wine-appreciation class, I asked her why she was paying money to become dis-satisfied with wines she currently enjoyed. This did not go over well, but I still think that from a standard maximize-subjective-utility perspective I had a point.
[Need I add that I now realize the ensuing fight was not, in fact, actually about wine or wine-appreciation, and that I was even more a jerk when I was younger than I am now?])  But I am intensely skeptical that just expanding the decision tree is at all going to be a productive way of coming to grips with this.]]></description>
<dc:subject>to:NB economics decision_theory epicycles_multiplication_thereof why_not_choose_to_be_happy endogenous_preferences transformative_treatments</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:4d61a2a1fbd8/</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:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:epicycles_multiplication_thereof"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:why_not_choose_to_be_happy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:endogenous_preferences"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:transformative_treatments"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.cambridge.org/core/journals/evolutionary-human-sciences/article/fitnessmaximizers-employ-pessimistic-probability-weighting-for-decisions-under-risk/FCF743180A566332C8AF9F7E7406AB43">
    <title>Fitness-maximizers employ pessimistic probability weighting for decisions under risk | Evolutionary Human Sciences | Cambridge Core</title>
    <dc:date>2020-12-16T19:47:18+00:00</dc:date>
    <link>https://www.cambridge.org/core/journals/evolutionary-human-sciences/article/fitnessmaximizers-employ-pessimistic-probability-weighting-for-decisions-under-risk/FCF743180A566332C8AF9F7E7406AB43</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The standard theory of rationality posits that agents order preferences according to average utilities associated with different choices. Expected utility theory has repeatedly failed as a predictive theory, as reflected in a growing literature in behavioural economics. Evolutionary theorists have suggested that seemingly irrational behaviours in contemporary contexts may have once served important functions, but existing work linking fitness and choice has not adequately addressed the challenges of constructing an evolutionary theory of decision making. In particular, fitness itself is not a reasonable metric for decision making since its timescale exceeds the lifespan of the decision-maker. Consequently, organisms use proximate systems that work on appropriate timescales and are amenable to feedback and learning. We develop an evolutionary principal–agent model in which individuals utilize a set of proximal choice variables to account for the non-linear dependence of these variables on consumption. While this is insufficient to maximize fitness in the presence of environmental stochasticity, maximum fitness can be achieved by adopting pessimistic probability weightings compatible with the rank-dependent expected utility family of choice models. In particular, pessimistic probability weighting emerges naturally in an evolutionary framework because of extreme intolerance to zeros in multiplicative growth processes."

--- Kelly gambling FTW?

]]></description>
<dc:subject>to:NB decision-making decision_theory evolutionary_biology evolution_of_cognition</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:375d11c1ccb8/</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-making"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:evolutionary_biology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:evolution_of_cognition"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://wwnorton.com/books/9781324004776/about-the-book/description">
    <title>Radical Uncertainty | John Kay, Mervyn King | W. W. Norton &amp; Company</title>
    <dc:date>2020-07-15T20:51:56+00:00</dc:date>
    <link>https://wwnorton.com/books/9781324004776/about-the-book/description</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Some uncertainties are resolvable. The insurance industry’s actuarial tables and the gambler’s roulette wheel both yield to the tools of probability theory. Most situations in life, however, involve a deeper kind of uncertainty, a radical uncertainty for which historical data provide no useful guidance to future outcomes. Radical uncertainty concerns events whose determinants are insufficiently understood for probabilities to be known or forecasting possible. Before President Barack Obama made the fateful decision to send in the Navy Seals, his advisers offered him wildly divergent estimates of the odds that Osama bin Laden would be in the Abbottabad compound. In 2000, no one—not least Steve Jobs—knew what a smartphone was; how could anyone have predicted how many would be sold in 2020? And financial advisers who confidently provide the information required in the standard retirement planning package—what will interest rates, the cost of living, and your state of health be in 2050?—demonstrate only that their advice is worthless.
"The limits of certainty demonstrate the power of human judgment over artificial intelligence. In most critical decisions there can be no forecasts or probability distributions on which we might sensibly rely. Instead of inventing numbers to fill the gaps in our knowledge, we should adopt business, political, and personal strategies that will be robust to alternative futures and resilient to unpredictable events. Within the security of such a robust and resilient reference narrative, uncertainty can be embraced, because it is the source of creativity, excitement, and profit."]]></description>
<dc:subject>to:NB books:noted decision_theory risk_vs_uncertainty popular_social_science</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:bfce5073cfe4/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:popular_social_science"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.cambridge.org/us/academic/subjects/economics/econometrics-statistics-and-mathematical-economics/ignorance-and-uncertainty?format=PB">
    <title>Ignorance and uncertainty | Econometrics, statistics and mathematical economics | Cambridge University Press</title>
    <dc:date>2020-01-30T23:56:30+00:00</dc:date>
    <link>https://www.cambridge.org/us/academic/subjects/economics/econometrics-statistics-and-mathematical-economics/ignorance-and-uncertainty?format=PB</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>to:NB books:noted decision_theory economics books:suggest_to_library</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:5a7f36f11016/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:books:suggest_to_library"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.nature.com/articles/s41567-019-0732-0">
    <title>The ergodicity problem in economics | Nature Physics</title>
    <dc:date>2019-12-06T17:57:25+00:00</dc:date>
    <link>https://www.nature.com/articles/s41567-019-0732-0</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[I find this intensely disappointing.
(I) We know, experimentally, that people do not maximize _ex ante_ expected utility.
(II) The decision rule that is supposed to be replacing "maximize expected utility" is, evidently, "maximize the growth rate of wealth", i.e., "maximize the long-run average increment to log wealth".
  A. So why isn't this is just re-inventing Kelly gambling?  (Yes, Kelly is cited.)
  B. For the long-run growth rate of wealth to be a uniquely-defined quantity  _presumes_ ergodicity for for the increments of log-wealth. If increments to log-wealth are stationary but not ergodic, this time average converges to a random quantity; if log wealth is an I(2) process (or higher-order integrated process), then it won't converge at all.  The rule "pick the option with the higher long-run time-average growth rate" is therefore insufficient, and we do not see what "ergodicity economics" predicts (or advises) in some very basic situations.  (These situations just happen not to include geometric Brownian motion.)
(III) Presuming that the time-average growth rate of wealth converges, the decision maker does not necessarily _know_ what it will converge _to_.  This is a fundamental problem of decision-making under _uncertainty_, as opposed to stochastic _risk_.  All of the examples presume the data-generating process is completely known to the decision-maker, which is an extremely strong form of rational expectations.  What does "ergodicity economics" predict about choosing between two GBMs with uncertainty about the growth rate?
]]></description>
<dc:subject>economics ergodic_theory decision_theory 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:1ed7cdb276ce/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:ergodic_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<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="https://arxiv.org/abs/1909.06853">
    <title>[1909.06853] Statistical inference for statistical decisions</title>
    <dc:date>2019-09-24T05:44:29+00:00</dc:date>
    <link>https://arxiv.org/abs/1909.06853</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The Wald development of statistical decision theory addresses decision making with sample data. Wald's concept of a statistical decision function (SDF) embraces all mappings of the form [data -> decision]. An SDF need not perform statistical inference; that is, it need not use data to draw conclusions about the true state of nature. Inference-based SDFs have the sequential form [data -> inference -> decision]. This paper motivates inference-based SDFs as practical procedures for decision making that may accomplish some of what Wald envisioned. The paper first addresses binary choice problems, where all SDFs may be viewed as hypothesis tests. It next considers as-if optimization, which uses a point estimate of the true state as if the estimate were accurate. It then extends this idea to as-if maximin and minimax-regret decisions, which use point estimates of some features of the true state as if they were accurate. The paper primarily uses finite-sample maximum regret to evaluate the performance of inference-based SDFs. To illustrate abstract ideas, it presents specific findings concerning treatment choice and point prediction with sample data."]]></description>
<dc:subject>to:NB decision_theory statistics to_read manski.charles_f.</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:b39b67dcce94/</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:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:manski.charles_f."/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1906.07801">
    <title>[1906.07801] Safe Testing</title>
    <dc:date>2019-06-23T17:11:02+00:00</dc:date>
    <link>https://arxiv.org/abs/1906.07801</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We present a new theory of hypothesis testing. The main concept is the S-value, a notion of evidence which, unlike p-values, allows for effortlessly combining evidence from several tests, even in the common scenario where the decision to perform a new test depends on the previous test outcome: safe tests based on S-values generally preserve Type-I error guarantees under such "optional continuation". S-values exist for completely general testing problems with composite null and alternatives. Their prime interpretation is in terms of gambling or investing, each S-value corresponding to a particular investment. Surprisingly, optimal "GROW" S-values, which lead to fastest capital growth, are fully characterized by the joint information projection (JIPr) between the set of all Bayes marginal distributions on H0 and H1. Thus, optimal S-values also have an interpretation as Bayes factors, with priors given by the JIPr. We illustrate the theory using two classical testing scenarios: the one-sample t-test and the 2x2 contingency table. In the t-test setting, GROW s-values correspond to adopting the right Haar prior on the variance, like in Jeffreys' Bayesian t-test. However, unlike Jeffreys', the "default" safe t-test puts a discrete 2-point prior on the effect size, leading to better behavior in terms of statistical power. Sharing Fisherian, Neymanian and Jeffreys-Bayesian interpretations, S-values and safe tests may provide a methodology acceptable to adherents of all three schools."]]></description>
<dc:subject>to:NB hypothesis_testing statistics information_theory decision_theory grunwald.peter kelly_gambling</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:00aba69e82c1/</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:hypothesis_testing"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:information_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:grunwald.peter"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kelly_gambling"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://arxiv.org/abs/1704.04118">
    <title>[1704.04118] From Data to Decisions: Distributionally Robust Optimization is Optimal</title>
    <dc:date>2019-06-17T22:15:31+00:00</dc:date>
    <link>https://arxiv.org/abs/1704.04118</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We study stochastic programs where the decision-maker cannot observe the distribution of the exogenous uncertainties but has access to a finite set of independent samples from this distribution. In this setting, the goal is to find a procedure that transforms the data to an estimate of the expected cost function under the unknown data-generating distribution, i.e., a predictor, and an optimizer of the estimated cost function that serves as a near-optimal candidate decision, i.e., a prescriptor. As functions of the data, predictors and prescriptors constitute statistical estimators. We propose a meta-optimization problem to find the least conservative predictors and prescriptors subject to constraints on their out-of-sample disappointment. The out-of-sample disappointment quantifies the probability that the actual expected cost of the candidate decision under the unknown true distribution exceeds its predicted cost. Leveraging tools from large deviations theory, we prove that this meta-optimization problem admits a unique solution: The best predictor-prescriptor pair is obtained by solving a distributionally robust optimization problem over all distributions within a given relative entropy distance from the empirical distribution of the data."

--- Physicists re-inventing learning theory for generalization error bounds?]]></description>
<dc:subject>to:NB to_read learning_theory large_deviations decision_theory statistics to_teach:childs_garden_of_statistical_learning_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:cc69f34c2117/</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:learning_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:large_deviations"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_teach:childs_garden_of_statistical_learning_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://danielcsaba.com/content/bayes_miss_Oct2018.pdf">
    <title>Learning with Misspecified Models</title>
    <dc:date>2018-12-13T05:10:16+00:00</dc:date>
    <link>http://danielcsaba.com/content/bayes_miss_Oct2018.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We consider Bayesian learning about a stable environment when the learner’s entertained probability
distributions (likelihoods) are all misspecified. We evaluate likelihoods according to the long-run average
payoff of the policy function they induce. We then show that, generically, the value that the Bayesian
learner attains in the long run is lower than what would be achievable with her misspecified set of likelihoods.
We introduce two kinds of indifference curves over the learner’s set: one based on the likelihoods’
induced long-run average payoff, and another capturing their statistical similarity. In case of misspecification,
we show that misalignment of these curves can lead the Bayesian learner to focus on payoff-irrelevant
features of the environment. On the other hand, under correct specification this misalignment has no bite.
We provide conditions under which it is feasible to construct an exponential family that allows the learner
to implement the best attainable policy in the long-run irrespective of misspecification. We demonstrate
applications of the introduced concepts through examples."]]></description>
<dc:subject>to:NB to_read bayesian_consistency decision_theory statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:00a9ce518d83/</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:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.pnas.org/content/115/44/E10387?etoc=">
    <title>Collective decision making by rational individuals | PNAS</title>
    <dc:date>2018-10-30T17:56:40+00:00</dc:date>
    <link>http://www.pnas.org/content/115/44/E10387?etoc=</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The patterns and mechanisms of collective decision making in humans and animals have attracted both empirical and theoretical attention. Of particular interest has been the variety of social feedback rules and the extent to which these behavioral rules can be explained and predicted from theories of rational estimation and decision making. However, models that aim to model the full range of social information use have incorporated ad hoc departures from rational decision-making theory to explain the apparent stochasticity and variability of behavior. In this paper I develop a model of social information use and collective decision making by fully rational agents that reveals how a wide range of apparently stochastic social decision rules emerge from fundamental information asymmetries both between individuals and between the decision makers and the observer of those decisions. As well as showing that rational decision making is consistent with empirical observations of collective behavior, this model makes several testable predictions about how individuals make decisions in groups and offers a valuable perspective on how we view sources of variability in animal, and human, behavior."]]></description>
<dc:subject>collective_cognition collective_action decision_theory re:democratic_cognition in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f0ceeebd885c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:collective_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:collective_action"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:democratic_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.jacobinmag.com/2015/09/vietnam-war-cambodia-ellsberg-pentagon-papers-kissinger">
    <title>War Without Reason</title>
    <dc:date>2018-08-24T14:40:28+00:00</dc:date>
    <link>https://www.jacobinmag.com/2015/09/vietnam-war-cambodia-ellsberg-pentagon-papers-kissinger</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>us_military american_hegemony irrationalism decision_theory moral_psychology kissinger.henry ellsberg.daniel have_read vietnam_war</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:72a54211deea/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:us_military"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:american_hegemony"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:irrationalism"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:moral_psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kissinger.henry"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:ellsberg.daniel"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:vietnam_war"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.aeaweb.org/articles?id=10.1257/aer.20161079">
    <title>The Design and Price of Information - American Economic Association</title>
    <dc:date>2018-01-24T23:23:17+00:00</dc:date>
    <link>https://www.aeaweb.org/articles?id=10.1257/aer.20161079</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A data buyer faces a decision problem under uncertainty. He can augment his initial private information with supplemental data from a data seller. His willingness to pay for supplemental data is determined by the quality of his initial private information. The data seller optimally offers a menu of statistical experiments. We establish the properties that any revenue-maximizing menu of experiments must satisfy. Every experiment is a non-dispersed stochastic matrix, and every menu contains a fully informative experiment. In the cases of binary states and actions, or binary types, we provide an explicit construction of the optimal menu of experiments."]]></description>
<dc:subject>to:NB decision_theory economics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:08a285c73955/</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:economics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://link.springer.com/article/10.1007/s11238-017-9600-5">
    <title>When and how to satisfice: an experimental investigation | SpringerLink</title>
    <dc:date>2017-10-13T00:38:23+00:00</dc:date>
    <link>https://link.springer.com/article/10.1007/s11238-017-9600-5</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This paper is about satisficing behaviour. Rather tautologically, this is when decision-makers are satisfied with achieving some objective, rather than in obtaining the best outcome. The term was coined by Simon (Q J Econ 69:99–118, 1955), and has stimulated many discussions and theories. Prominent amongst these theories are models of incomplete preferences, models of behaviour under ambiguity, theories of rational inattention, and search theories. Most of these, however, seem to lack an answer to at least one of two key questions: when should the decision-maker (DM) satisfice; and how should the DM satisfice. In a sense, search models answer the latter question (in that the theory tells the DM when to stop searching), but not the former; moreover, usually the question as to whether any search at all is justified is left to a footnote. A recent paper by Manski (Theory Decis. doi:10.1007/s11238-017-9592-1, 2017) fills the gaps in the literature and answers the questions: when and how to satisfice? He achieves this by setting the decision problem in an ambiguous situation (so that probabilities do not exist, and many preference functionals can therefore not be applied) and by using the Minimax Regret criterion as the preference functional. The results are simple and intuitive. This paper reports on an experimental test of his theory. The results show that some of his propositions (those relating to the ‘how’) appear to be empirically valid while others (those relating to the ‘when’) are less so."

--- I am continually impressed by economist's resistance to Simon's (very simple) points about computational complexity.  (But: two cheers for actually doing something empirical.)]]></description>
<dc:subject>to:NB decision-making decision_theory bounded_rationality</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1e9e1fccbc84/</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-making"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bounded_rationality"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.pnas.org/content/114/32/8499.abstract.html">
    <title>Optimal decision making and matching are tied through diminishing returns</title>
    <dc:date>2017-08-08T17:13:00+00:00</dc:date>
    <link>http://www.pnas.org/content/114/32/8499.abstract.html</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["How individuals make decisions has been a matter of long-standing debate among economists and researchers in the life sciences. In economics, subjects are viewed as optimal decision makers who maximize their overall reward income. This framework has been widely influential, but requires a complete knowledge of the reward contingencies associated with a given choice situation. Psychologists and ecologists have observed that individuals tend to use a simpler “matching” strategy, distributing their behavior in proportion to relative rewards associated with their options. This article demonstrates that the two dominant frameworks of choice behavior are linked through the law of diminishing returns. The relatively simple matching can in fact provide maximal reward when the rewards associated with decision makers’ options saturate with the invested effort. Such saturating relationships between reward and effort are hallmarks of the law of diminishing returns. Given the prevalence of diminishing returns in nature and social settings, this finding can explain why humans and animals so commonly behave according to the matching law. The article underscores the importance of the law of diminishing returns in choice behavior."]]></description>
<dc:subject>to:NB decision-making decision_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:87b86cedc8af/</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-making"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://link.springer.com/article/10.1007/s11238-017-9592-1?wt_mc=alerts.TOCjournals">
    <title>Optimize, satisfice, or choose without deliberation? A simple minimax-regret assessment | SpringerLink</title>
    <dc:date>2017-07-31T16:02:18+00:00</dc:date>
    <link>https://link.springer.com/article/10.1007/s11238-017-9592-1?wt_mc=alerts.TOCjournals</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Simon (Q J Econ 69:99–118, 1955) introduced satisficing, but he did not provide a precise definition or analysis. Other researchers have subsequently interpreted satisficing in various ways, but a consensus perspective still has not emerged. This paper interprets satisficing as a class of decision strategies that a person might use when seeking to optimize in a setting where deliberation is costly. Costly deliberation lies at the heart of Simon’s motivation of satisficing, but he did not formalize the idea. I do so here, studying decision making as a problem of minimax-regret planning in which costly deliberation enables a person to reduce ambiguity. I report simple specific findings on how the magnitude of deliberation costs may affect choice of a decision strategy."]]></description>
<dc:subject>to:NB decision-making decision_theory bounded_rationality manski.charles_f.</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:081bad7b9eb8/</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-making"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bounded_rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:manski.charles_f."/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.press.uchicago.edu/ucp/books/book/chicago/R/bo3639393.html">
    <title>Rationalizing Capitalist Democracy: The Cold War Origins of Rational Choice Liberalism, Amadae</title>
    <dc:date>2017-07-10T07:12:41+00:00</dc:date>
    <link>http://www.press.uchicago.edu/ucp/books/book/chicago/R/bo3639393.html</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In Rationalizing Capitalist Democracy, S. M. Amadae tells the remarkable story of how rational choice theory rose from obscurity to become the intellectual bulwark of capitalist democracy. Amadae roots Rationalizing Capitalist Democracy in the turbulent post-World War II era, showing how rational choice theory grew out of the RAND Corporation's efforts to develop a "science" of military and policy decisionmaking. But while the first generation of rational choice theorists—William Riker, Kenneth Arrow, and James Buchanan—were committed to constructing a "scientific" approach to social science research, they were also deeply committed to defending American democracy from its Marxist critics. Amadae reveals not only how the ideological battles of the Cold War shaped their ideas but also how those ideas may today be undermining the very notion of individual liberty they were created to defend."]]></description>
<dc:subject>books:noted to:NB history_of_science history_of_ideas political_philosophy economics political_economy cold_war game_theory decision_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:d9cfdc5b10a0/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:books:noted"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:history_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:history_of_ideas"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:political_philosophy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:political_economy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cold_war"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:game_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
</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="https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2729537">
    <title>Active Learning with Misspecified Beliefs by Drew Fudenberg , Gleb Romanyuk, Philipp Strack :: SSRN</title>
    <dc:date>2016-12-13T14:40:44+00:00</dc:date>
    <link>https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2729537</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We study learning and information acquisition by a Bayesian agent who is misspecified in the sense that his prior belief assigns probability zero to the true state of the world. In our model, at each instant the agent takes an action and observes the corresponding payoff, which is the sum of the payoff generated by a fixed but unknown function and an additive error term. We provide a complete characterization of asymptotic actions and beliefs when the agent's subjective state space is a doubleton. A simple example with three actions shows that in a misspecified environment a myopic agent's beliefs converge while a sufficiently patient agent's beliefs do not. This shows that examples of myopic agents with non-converging beliefs in the prior literature require all myopically optimal actions to be informative, and illustrates a novel interaction between misspecification and the agent's subjective interest rate."]]></description>
<dc:subject>to:NB statistics bayesian_consistency misspecification decision_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:faa7a13edadd/</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:bayesian_consistency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1007/s11229-015-0846-6?wt_mc=alerts.TOCjournals">
    <title>Belief without credence | SpringerLink</title>
    <dc:date>2016-09-08T20:49:36+00:00</dc:date>
    <link>http://link.springer.com/article/10.1007/s11229-015-0846-6?wt_mc=alerts.TOCjournals</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["One of the deepest ideological divides in contemporary epistemology concerns the relative importance of belief versus credence. A prominent consideration in favor of credence-based epistemology is the ease with which it appears to account for rational action. In contrast, cases with risky payoff structures threaten to break the link between rational belief and rational action. This threat poses a challenge to traditional epistemology, which maintains the theoretical prominence of belief. The core problem, we suggest, is that belief may not be enough to register all aspects of a subject’s epistemic position with respect to any given proposition. We claim this problem can be solved by introducing other doxastic attitudes—genuine representations—that differ in strength from belief. The resulting alternative picture, a kind of doxastic states pluralism, retains the central features of traditional epistemology—most saliently, an emphasis on truth as a kind of objective accuracy—while adequately accounting for rational action."]]></description>
<dc:subject>to:NB epistemology decision_theory rationality</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:3b0ea103ed79/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:rationality"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.aeaweb.org/articles?id=10.1257/aer.20150678">
    <title>Robust Social Decisions</title>
    <dc:date>2016-08-30T21:57:54+00:00</dc:date>
    <link>https://www.aeaweb.org/articles?id=10.1257/aer.20150678</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We propose and operationalize normative principles to guide social decisions when individuals potentially have imprecise and heterogeneous beliefs, in addition to conflicting tastes or interests. To do so, we adapt the standard Pareto principle to those preference comparisons that are robust to belief imprecision and characterize social preferences that respect this robust principle. We also characterize a suitable restriction of this principle. The former principle provides stronger guidance when it can be satisfied; when it cannot, the latter always provides minimal guidance."]]></description>
<dc:subject>to:NB decision_theory social_choice risk_vs_uncertainty re:knightian_uncertainty</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:818e0cbacb75/</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:social_choice"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:knightian_uncertainty"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1007/s11238-015-9526-8">
    <title>Minimizing regret in dynamic decision problems - Springer</title>
    <dc:date>2016-05-31T14:25:55+00:00</dc:date>
    <link>http://link.springer.com/article/10.1007/s11238-015-9526-8</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The menu-dependent nature of regret minimization creates subtleties when it is applied to dynamic decision problems. It is not clear whether forgone opportunities should be included in the menu. We explain commonly observed behavioral patterns as minimizing regret when forgone opportunities are present. If forgone opportunities are included, we can characterize when a form of dynamic consistency is guaranteed."]]></description>
<dc:subject>decision_theory low-regret_learning halpern.joseph_y. in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:aedebf5be7d2/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:low-regret_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:halpern.joseph_y."/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://nostalgebraist.tumblr.com/post/143718406034/the-future-of-humanity-institute-seems-very">
    <title>trees are harlequins, words are harlequins | the future of humanity institute seems very...</title>
    <dc:date>2016-05-04T02:54:39+00:00</dc:date>
    <link>http://nostalgebraist.tumblr.com/post/143718406034/the-future-of-humanity-institute-seems-very</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>decision_theory utter_stupidity</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:b8119e6a570c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:utter_stupidity"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://orgtheory.wordpress.com/2015/06/26/picking-the-right-metric-from-college-ratings-to-the-cold-war/#more-29713">
    <title>picking the right metric: from college ratings to the cold war | orgtheory.net</title>
    <dc:date>2015-07-11T02:53:15+00:00</dc:date>
    <link>https://orgtheory.wordpress.com/2015/06/26/picking-the-right-metric-from-college-ratings-to-the-cold-war/#more-29713</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>have_read decision_theory operations_research optimization economics but_what_do_i_optimize history_of_science RAND single_vision_and_newtons_sleep to:blog</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:3b396413a4b6/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:operations_research"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:but_what_do_i_optimize"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:history_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:RAND"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:single_vision_and_newtons_sleep"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://laeuferpaar.de/Papers/ExpertsGroupDecisions.pdf">
    <title>Modeling Individual Expertise in Group Judgments</title>
    <dc:date>2014-12-04T15:25:04+00:00</dc:date>
    <link>http://laeuferpaar.de/Papers/ExpertsGroupDecisions.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Group judgments are often—implicitly or explicitly—influenced by their members’ individual expertise. However, given that expertise is seldom recognized fully and that some distortions may occur (bias, correlation, etc.), it is not clear that differential weighting is an epistemically advantageous strategy with respect to straight averaging. Our paper characterizes a wide set of conditions under which differential weighting outperforms straight averaging and embeds the results into the multidisciplinary group decision-making literature."]]></description>
<dc:subject>to_read collective_cognition re:democratic_cognition decision_theory in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e42e4b9ea2d8/</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:collective_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:democratic_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://mitpress.mit.edu/books/rational-action">
    <title>Rational Action | The MIT Press</title>
    <dc:date>2014-11-25T02:19:45+00:00</dc:date>
    <link>http://mitpress.mit.edu/books/rational-action</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["During World War II, the Allied military forces faced severe problems integrating equipment, tactics, and logistics into successful combat operations. To help confront these problems, scientists and engineers developed new means of studying which equipment designs would best meet the military’s requirements and how the military could best use the equipment it had on hand. By 1941 they had also begun to gather and analyze data from combat operations to improve military leaders’ ordinary planning activities. In Rational Action, William Thomas details these developments, and how they gave rise during the 1950s to a constellation of influential new fields—which he terms the “sciences of policy”—that included operations research, management science, systems analysis, and decision theory.
"Proponents of these new sciences embraced a variety of agendas. Some aimed to improve policymaking directly, while others theorized about how one decision could be considered more rational than another. Their work spanned systems engineering, applied mathematics, nuclear strategy, and the philosophy of science, and it found new niches in universities, in businesses, and at think tanks such as the RAND Corporation. The sciences of policy also took a prominent place in epic narratives told about the relationships among science, state, and society in an intellectual culture preoccupied with how technology and reason would shape the future. Thomas follows all these threads to illuminate and make new sense of the intricate relationships among scientific analysis, policymaking procedure, and institutional legitimacy at a crucial moment in British and American history."]]></description>
<dc:subject>books:noted decision_theory statistics economics history_of_science cold_war WWII optimization in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:4921051f048e/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:books:noted"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:history_of_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cold_war"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:WWII"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="https://www.aeaweb.org/articles.php?doi=10.1257/jep.28.4.213">
    <title>JEP (28,4) p. 213 - Retrospectives: The Cold-War Origins of the Value of Statistical Life</title>
    <dc:date>2014-11-13T01:33:43+00:00</dc:date>
    <link>https://www.aeaweb.org/articles.php?doi=10.1257/jep.28.4.213</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This paper traces the history of the "Value of Statistical Life" (VSL), which today is used routinely in benefit-cost analysis of life-saving investments. The "value of statistical life" terminology was introduced by Thomas Schelling (1968) in his essay, "The Life You Save May Be Your Own." Schelling made the crucial move to think in terms of risk rather than individual lives, with the hope to dodge the moral thicket of valuing "life." But as recent policy debates have illustrated, his move only thickened it. Tellingly, interest in the subject can be traced back another twenty years before Schelling's essay to a controversy at RAND Corporation following its earliest application of operations research to defense planning. RAND wanted to avoid valuing pilot's lives but the Air Force insisted they confront the issue. Thus, the VSL is not only well acquainted with political controversy; it was born from it."]]></description>
<dc:subject>decision_theory economics moral_philosophy cold_war schelling.thomas cost-benefit_analysis asking_the_egg_what_it_would_give_to_not_be_in_the_omlet in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f1c5a600091a/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:moral_philosophy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cold_war"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:schelling.thomas"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cost-benefit_analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:asking_the_egg_what_it_would_give_to_not_be_in_the_omlet"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://cowles.econ.yale.edu/P/cp/p10b/p1053.pdf">
    <title>Knightian Decision Theory. Part I</title>
    <dc:date>2014-08-01T18:33:53+00:00</dc:date>
    <link>http://cowles.econ.yale.edu/P/cp/p10b/p1053.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A theory of choice under uncertainty is proposed which removes the completeness assumption from the Anscombe–Aumann formulation of Savage’s theory and introduces an inertia assumption. The inertia assumption is that there is such a thing as the status quo and an alternative is accepted only if it is preferred to the status quo. This theory is one way of giving rigorous expression to Frank Knight’s distinction between risk and uncertainty."

--- The inertia assumption rules out money pumps in the form of trading cycles, but I don't see how it evades Dutch Books, which are a set of _compound_ bets at odds guaranteed to cost the agent money.]]></description>
<dc:subject>decision_theory risk_vs_uncertainty economics re:knightian_uncertainty bewley.truman in_NB have_read to:blog via:mraginsky</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:50ffc098f332/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:knightian_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bewley.truman"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<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:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:mraginsky"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://statweb.stanford.edu/~cgates/PERSI/papers/thinking.pdf">
    <title>Thinking Too Much (Diaconis)</title>
    <dc:date>2014-07-22T17:31:46+00:00</dc:date>
    <link>http://statweb.stanford.edu/~cgates/PERSI/papers/thinking.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[Includes the " 'why not use decision theory, like you teach?' 'Come on, this is serious' " exchange.  But also looks worth reading on broader grounds.]]></description>
<dc:subject>to_read decision_theory bounded_rationality rationality epistemology diaconis.persi foundations_of_statistics to:blog</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1d372229f62e/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bounded_rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:epistemology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:diaconis.persi"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:foundations_of_statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to:blog"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.slate.com/articles/technology/bitwise/2014/07/roko_s_basilisk_the_most_terrifying_thought_experiment_of_all_time.single.html">
    <title>Roko’s Basilisk: The most terrifying thought experiment of all time.</title>
    <dc:date>2014-07-19T14:43:24+00:00</dc:date>
    <link>http://www.slate.com/articles/technology/bitwise/2014/07/roko_s_basilisk_the_most_terrifying_thought_experiment_of_all_time.single.html</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>utter_stupidity decision_theory the_rapture_for_nerds moral_philosophy psychoceramics auerbach.david blogged lesswrong impressive_act_what_do_you_call_yourselves_the_rationalists</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c0cf310af56c/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:utter_stupidity"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:the_rapture_for_nerds"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:moral_philosophy"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:psychoceramics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:auerbach.david"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:blogged"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:lesswrong"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:impressive_act_what_do_you_call_yourselves_the_rationalists"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1406.7563">
    <title>[1406.7563] When is a crowd wise?</title>
    <dc:date>2014-07-12T01:02:51+00:00</dc:date>
    <link>http://arxiv.org/abs/1406.7563</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Numerous studies and anecdotes demonstrate the "wisdom of the crowd," the surprising accuracy of a group's aggregated judgments. Less is known, however, about the generality of crowd wisdom. For example, are crowds wise even if their members have systematic judgmental biases, or can influence each other before members render their judgments? If so, are there situations in which we can expect a crowd to be less accurate than skilled individuals? We provide a precise but general definition of crowd wisdom: A crowd is wise if a linear aggregate, for example a mean, of its members' judgments is closer to the target value than a randomly, but not necessarily uniformly, sampled member of the crowd. Building on this definition, we develop a theoretical framework for examining, a priori, when and to what degree a crowd will be wise. We systematically investigate the boundary conditions for crowd wisdom within this framework and determine conditions under which the accuracy advantage for crowds is maximized. Our results demonstrate that crowd wisdom is highly robust: Even if judgments are biased and correlated, one would need to nearly deterministically select only a highly skilled judge before an individual's judgment could be expected to be more accurate than a simple averaging of the crowd. Our results also provide an accuracy rationale behind the need for diversity of judgments among group members. Contrary to folk explanations of crowd wisdom which hold that judgments should ideally be independent so that errors cancel out, we find that crowd wisdom is maximized when judgments systematically differ as much as possible. We re-analyze data from two published studies that confirm our theoretical results."

--- Compare to gated version at http://psycnet.apa.org/psycinfo/2014-03872-001/]]></description>
<dc:subject>to_read collective_cognition re:democratic_cognition prediction decision_theory via:henry_farrell in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:2ead13c2c666/</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:collective_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:democratic_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:henry_farrell"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1007/s11238-013-9393-0">
    <title>Axiomatizing bounded rationality: the priority heuristic - Springer</title>
    <dc:date>2014-07-08T03:22:38+00:00</dc:date>
    <link>http://link.springer.com/article/10.1007/s11238-013-9393-0</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["This paper presents an axiomatic framework for the priority heuristic, a model of bounded rationality in Selten’s (in: Gigerenzer and Selten (eds.) Bounded rationality: the adaptive toolbox, 2001) spirit of using empirical evidence on heuristics. The priority heuristic predicts actual human choices between risky gambles well. It implies violations of expected utility theory such as common consequence effects, common ratio effects, the fourfold pattern of risk taking and the reflection effect. We present an axiomatization of a parameterized version of the heuristic which generalizes the heuristic in order to account for individual differences and inconsistencies. The axiomatization uses semiorders (Luce, Econometrica 24:178–191, 1956), which have an intransitive indifference part and a transitive strict preference component. The axiomatization suggests new testable predictions of the priority heuristic and makes it easier for theorists to study the relation between heuristics and other axiomatic theories such as cumulative prospect theory."]]></description>
<dc:subject>to:NB heuristics decision_theory cognitive_science gigerenzer.gerd</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:7016a155d2ee/</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:heuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_science"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:gigerenzer.gerd"/>
</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://journal.sjdm.org/14/14206/jdm14206.html">
    <title>Approximating rationality under incomplete information: Adaptive inferences for missing cue values based on cue-discrimination</title>
    <dc:date>2014-03-31T19:00:28+00:00</dc:date>
    <link>http://journal.sjdm.org/14/14206/jdm14206.html</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In a highly uncertain world, individuals often have to make decisions in situations with incomplete information. We investigated in three experiments how partial cue information is treated in complex probabilistic inference tasks. Specifically, we test a mechanism to infer missing cue values that is based on the discrimination rate of cues (i.e., how often a cue makes distinct predictions for choice options). We show analytically that inferring missing cue values based on discrimination rate maximizes the probability for a correct inference in many decision environments and that it is therefore adaptive to use it. Results from three experiments show that individuals are sensitive to the discrimination rate and use it when it is a valid inference mechanism but rely on other inference mechanisms, such as the cues’ base-rate of positive information, when it is not. We find adaptive inferences for incomplete information in environments in which participants are explicitly provided with information concerning the base-rate and discrimination rate of cues (Exp. 1) as well as in environments in which they learn these properties by experience (Exp. 2). Results also hold in environments of further increased complexity (Exp. 3). In all studies, participants show a high ability to adaptively infer incomplete information and to integrate this inferred information with other available cues to approximate the naïve Bayesian solution."]]></description>
<dc:subject>to:NB experimental_psychology decision_theory heuristics cognitive_science</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:931d2c25a5b2/</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:experimental_psychology"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:heuristics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cognitive_science"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1402.6118">
    <title>[1402.6118] Approximate Models and Robust Decisions</title>
    <dc:date>2014-03-08T22:28:11+00:00</dc:date>
    <link>http://arxiv.org/abs/1402.6118</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Statistical decisions based partly or solely on predictions from probabilistic models may be sensitive to model misspecification. Statisticians are taught from an early stage that ``all models are wrong'' but little formal guidance exists on how to assess the impact of model approximation, or how to proceed when optimal actions appear sensitive to model fidelity. This article presents one potential applied framework to address this. We discuss diagnostic techniques, including graphical approaches and summary statistics, to help highlight decisions made through minimised expected loss that are sensitive to model misspecification. We then derive formal methods for decision making under model misspecification by quantifying stability of optimal actions to perturbations within a neighbourhood of model space, defined via an information (Kullback-Leibler) divergence around the approximating model. This latter approach draws heavily from recent work in the robust control, macroeconomics and financial mathematics literature. We adopt a Bayesian approach throughout although the methods are agnostic to this position."]]></description>
<dc:subject>to:NB robustness misspecification decision_theory statistics</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e2ba3b28fa4a/</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:robustness"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:misspecification"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1007/s11229-014-0398-1">
    <title>On the regress problem of deciding how to decide - Springer</title>
    <dc:date>2014-02-24T17:43:31+00:00</dc:date>
    <link>http://link.springer.com/article/10.1007/s11229-014-0398-1</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Any decision is made in some way or another. Which way? (Have I worked out enough alternatives to choose from? Which decision rule to apply?) That is a higher-order decision problem, to be dealt with in some way or other. Which way? That is an even higher-order decision problem. There seems to be a regress of decision problems toward higher and higher orders. But in daily life we stop moving to higher-order decision problems—stop the regress—at some finite point. The regress problem of deciding how to decide is the problem of explaining what would make it rational to stop the regress. I will give a new solution in the present paper. The result suggests a new way of looking at standard Bayesian theory and the more recent theory of adaptive rationality."
]]></description>
<dc:subject>to:NB rationality decision_theory have_skimmed</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:5afecf36b5a3/</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:rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_skimmed"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1402.4884">
    <title>[1402.4884] Le Cam meets LeCun: Deficiency and Generic Feature Learning</title>
    <dc:date>2014-02-21T19:18:55+00:00</dc:date>
    <link>http://arxiv.org/abs/1402.4884</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[""Deep Learning" methods attempt to learn generic features in an unsupervised fashion from a large unlabelled data set. These generic features should perform as well as the best hand crafted features for any learning problem that makes use of this data. We provide a definition of generic features, characterize when it is possible to learn them and provide methods closely related to the autoencoder and deep belief network of deep learning. In order to do so we use the notion of deficiency and illustrate its value in studying certain general learning problems."]]></description>
<dc:subject>to_read statistics learning_theory sufficiency decision_theory in_NB entableted</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:c0b08809acbb/</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:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:learning_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:sufficiency"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:entableted"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://global.oup.com/academic/product/risk-and-rationality-9780199672165">
    <title>Risk and Rationality - Lara Buchak - Oxford University Press</title>
    <dc:date>2014-02-20T21:05:29+00:00</dc:date>
    <link>http://global.oup.com/academic/product/risk-and-rationality-9780199672165</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Lara Buchak sets out an original account of the principles that govern rational decision-making in the face of risk. A distinctive feature of these decisions is that individuals are forced to consider how their choices will turn out under various circumstances, and decide how to trade off the possibility that a choice will turn out well against the possibility that it will turn out poorly. The orthodox view is that there is only one acceptable way to do this: rational individuals must maximize expected utility. Buchak's contention, however, is that the orthodox theory (expected utility theory) dictates an overly narrow way in which considerations about risk can play a role in an individual's choices. Combining research from economics and philosophy, she argues for an alternative, more permissive, theory of decision-making: one that allows individuals to pay special attention to the worst-case or best-case scenario (among other 'global features' of gambles). This theory, risk-weighted expected utility theory, better captures the preferences of actual decision-makers. Furthermore, it isolates the distinct roles that beliefs, desires, and risk-attitudes play in decision-making. Finally, contra the orthodox view, Buchak argues that decision-makers whose preferences can be captured by risk-weighted expected utility theory are rational. Thus, Risk and Rationality is in many ways a vindication of the ordinary decision-maker--particularly his or her attitude towards risk--from the point of view of even ideal rationality."]]></description>
<dc:subject>books:noted rationality decision_theory risk_vs_uncertainty in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:a8564f976938/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:books:noted"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-022513-115553">
    <title>Dynamic Treatment Regimes - Annual Review of Statistics and Its Application, 1(1):447</title>
    <dc:date>2014-01-16T00:01:55+00:00</dc:date>
    <link>http://www.annualreviews.org/doi/abs/10.1146/annurev-statistics-022513-115553</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["A dynamic treatment regime consists of a sequence of decision rules, one per stage of intervention, that dictate how to individualize treatments to patients, based on evolving treatment and covariate history. These regimes are particularly useful for managing chronic disorders and fit well into the larger paradigm of personalized medicine. They provide one way to operationalize a clinical decision support system. Statistics plays a key role in the construction of evidence-based dynamic treatment regimes—informing the best study design as well as efficient estimation and valid inference. Owing to the many novel methodological challenges this area offers, it has been growing in popularity among statisticians in recent years. In this article, we review the key developments in this exciting field of research. In particular, we discuss the sequential multiple assignment randomized trial designs, estimation techniques like Q-learning and marginal structural models, and several inference techniques designed to address the associated nonstandard asymptotics. We reference software whenever available. We also outline some important future directions."]]></description>
<dc:subject>decision_theory sequential_optimization reinforcement_learning causal_inference graphical_models statistics murphy.susan in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1a9b6828e858/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:sequential_optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:reinforcement_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:causal_inference"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:graphical_models"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:murphy.susan"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1312.1277">
    <title>[1312.1277] Bandits and Experts in Metric Spaces</title>
    <dc:date>2014-01-02T17:56:44+00:00</dc:date>
    <link>http://arxiv.org/abs/1312.1277</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["In a multi-armed bandit problem, an online algorithm chooses from a set of strategies in a sequence of trials so as to maximize the total payoff of the chosen strategies. While the performance of bandit algorithms with a small finite strategy set is quite well understood, bandit problems with large strategy sets are still a topic of very active investigation, motivated by practical applications such as online auctions and web advertisement. The goal of such research is to identify broad and natural classes of strategy sets and payoff functions which enable the design of efficient solutions. 
"In this work we study a very general setting for the multi-armed bandit problem in which the strategies form a metric space, and the payoff function satisfies a Lipschitz condition with respect to the metric. We refer to this problem as the "Lipschitz MAB problem". We present a solution for the multi-armed bandit problem in this setting. That is, for every metric space we define an isometry invariant which bounds from below the performance of Lipschitz MAB algorithms for this metric space, and we present an algorithm which comes arbitrarily close to meeting this bound. Furthermore, our technique gives even better results for benign payoff functions. We also address the full-feedback ("best expert") version of the problem, where after every round the payoffs from all arms are revealed."]]></description>
<dc:subject>to:NB low-regret_learning bandit_problems optimization decision_theory re:growing_ensemble_project</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1b8cc839aa66/</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:low-regret_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bandit_problems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:growing_ensemble_project"/>
</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://asset2013.org/archivo/ficheros/Microeconomics/19-CUHADAROGLU-TUGCE.pdf">
    <title>Choose what you like or like what you choose? Identifying Influence and Homophily out of Individual Decisions</title>
    <dc:date>2013-11-18T13:59:56+00:00</dc:date>
    <link>http://asset2013.org/archivo/ficheros/Microeconomics/19-CUHADAROGLU-TUGCE.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We investigate the microfoundations of the identification problem related to social influence and homophily. Focusing on the individual decision making of interacting individuals, we investigate how they affect each other’s behaviors. We propose simple and direct measures of homophily and influence by making use of individual preferences of these interacting individuals. Since in many occasions, preferences are not easily observed, we extend our analysis to the observables, decision outcomes. In order to infer the underlying preferences of interacting individuals out of their decision outcomes, we follow a foundational approach. We analyze the behavioral characteristics of individual de- cision making that includes interaction and finally we make use of
the tools that are provided by revealed preference theory in order to uncover the underlying preferences of the individuals. Based on re- vealed preference analysis, we revisit our measurement techniques for homophily and influence."

- Gets only partial identification.]]></description>
<dc:subject>to:NB to_read decision_theory social_influence re:homophily_and_confounding</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:34b0356b982e/</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:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:social_influence"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:homophily_and_confounding"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.pnas.org/content/110/39/E3704.abstract">
    <title>Decisions on the fly in cellular sensory systems</title>
    <dc:date>2013-10-21T15:45:36+00:00</dc:date>
    <link>http://www.pnas.org/content/110/39/E3704.abstract</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Cell-signaling pathways are often presumed to convert just the level of an external stimulus to response. However, in contexts such as the immune system or rapidly developing embryos, cells plausibly have to make rapid decisions based on limited information. Statistical theory defines absolute bounds on the minimum average observation time necessary for decisions subject to a defined error rate. We show that common genetic circuits have the potential to approach the theoretical optimal performance. They operate by accumulating a single chemical species and then applying a threshold. The circuit parameters required for optimal performance can be learned by a simple hill-climbing search. The complex but reversible protein modifications that accompany signaling thus have the potential to perform analog computations."]]></description>
<dc:subject>to:NB biochemical_networks decision_theory signal_transduction biological_computation biology</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1d93863b9998/</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:biochemical_networks"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:signal_transduction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:biological_computation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:biology"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1307.0473">
    <title>[1307.0473] Online discrete optimization in social networks in the presence of Knightian uncertainty</title>
    <dc:date>2013-07-02T03:17:00+00:00</dc:date>
    <link>http://arxiv.org/abs/1307.0473</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We study a model of collective real-time decision-making (or learning) in a social network operating in an uncertain environment, for which no a priori probabilistic model is available. Instead, the environment's impact on the agents in the network is seen through a sequence of cost functions, revealed to the agents in a causal manner only after all the relevant actions are taken. There are two kinds of costs: individual costs incurred by each agent and local-interaction costs incurred by each agent and its neighbors in the social network. Moreover, agents have inertia: each agent has a default mixed strategy that stays fixed regardless of the state of the environment, and must expend effort to deviate from this strategy in order to respond to cost signals coming from the environment. We construct a decentralized strategy, wherein each agent selects its action based only on the costs directly affecting it and on the decisions made by its neighbors in the network. In this setting, we quantify social learning in terms of regret, which is given by the difference between the realized network performance over a given time horizon and the best performance that could have been achieved in hindsight by a fictitious centralized entity with full knowledge of the environment's evolution. We show that our strategy achieves the regret that scales sublinearly with the time horizon and polylogarithmically with the number of agents and the maximum number of neighbors of any agent in the social network."]]></description>
<dc:subject>distributed_systems learning_theory low-regret_learning risk_vs_uncertainty social_life_of_the_mind kith_and_kin raginsky.maxim re:democratic_cognition in_NB decision_theory re:knightian_uncertainty have_read markov_models</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:27a152dd8649/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:distributed_systems"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:learning_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:low-regret_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:risk_vs_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:social_life_of_the_mind"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kith_and_kin"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:raginsky.maxim"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:democratic_cognition"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:knightian_uncertainty"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:have_read"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:markov_models"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.aeaweb.org/articles.php?doi=10.1257/jel.51.2.496">
    <title>Bounded-Rationality Models: Tasks to Become Intellectually Competitive</title>
    <dc:date>2013-06-23T00:10:13+00:00</dc:date>
    <link>http://www.aeaweb.org/articles.php?doi=10.1257/jel.51.2.496</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Research in experimental economics has cogently challenged the fundamental precept of neoclassical economics that economic agents optimize. The last two decades have seen elaboration of boundedly rational models that try to move away from the optimization approach, in ways consistent with experimental findings. Nonetheless, the collection of alternative models has made little headway supplanting the dominant paradigm. We delineate key ways in which neoclassical microeconomics holds continuing and compelling advantages over bounded-rationality models, and suggest, via a few examples, the sorts of further, difficult pushes that would be needed to redress this state of affairs. Closer collaboration between theoretic modeling and experiments is clearly seen to be necessary."]]></description>
<dc:subject>to:NB economics bounded_rationality decision_theory decision-making selten.reinhard social_science_methodology</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:f09dd742d2af/</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:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bounded_rationality"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision-making"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:selten.reinhard"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:social_science_methodology"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://link.springer.com/article/10.1023%2FA%3A1022899518027">
    <title>Variance and Bias for General Loss Functions - Springer</title>
    <dc:date>2013-06-10T18:49:29+00:00</dc:date>
    <link>http://link.springer.com/article/10.1023%2FA%3A1022899518027</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["When using squared error loss, bias and variance and their decomposition of prediction error are well understood and widely used concepts. However, there is no universally accepted definition for other loss functions. Numerous attempts have been made to extend these concepts beyond squared error loss. Most approaches have focused solely on 0-1 loss functions and have produced significantly different definitions. These differences stem from disagreement as to the essential characteristics that variance and bias should display. This paper suggests an explicit list of rules that we feel any “reasonable” set of definitions should satisfy. Using this framework, bias and variance definitions are produced which generalize to any symmetric loss function. We illustrate these statistics on several loss functions with particular emphasis on 0-1 loss. We conclude with a discussion of the various definitions that have been proposed in the past as well as a method for estimating these quantities on real data sets."

Ungated version (apparently): http://www-bcf.usc.edu/~gareth/research/newbv2.pdf]]></description>
<dc:subject>to:NB to_read bias-variance to_teach:undergrad-ADA to_teach:data-mining prediction decision_theory</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:75a5102c9d5c/</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:bias-variance"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_teach:undergrad-ADA"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:to_teach:data-mining"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://noahpinionblog.blogspot.com/2013/05/bets-do-not-necessarily-reveal-beliefs.html">
    <title>Noahpinion: Bets do not (necessarily) reveal beliefs</title>
    <dc:date>2013-05-27T17:02:36+00:00</dc:date>
    <link>http://noahpinionblog.blogspot.com/2013/05/bets-do-not-necessarily-reveal-beliefs.html</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[Well, yes.  (See also the idiocy of prediction markets.)]]></description>
<dc:subject>decision_theory gives_economists_a_bad_name evisceration prediction</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:da2bc29209d0/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:gives_economists_a_bad_name"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:evisceration"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:prediction"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.mcsweeneys.net/articles/its-not-you-its-quantitative-cost-benefit-analysis#.URO1r9JCDZA.twitter">
    <title>McSweeney’s Internet Tendency: It’s Not You, It’s Quantitative Cost-Benefit Analysis.</title>
    <dc:date>2013-02-07T17:13:35+00:00</dc:date>
    <link>http://www.mcsweeneys.net/articles/its-not-you-its-quantitative-cost-benefit-analysis#.URO1r9JCDZA.twitter</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>decision_theory funny:malicious funny:because_its_true practices_relating_to_the_transmission_of_genetic_information via:kjhealy</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:02b64af71073/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:funny:malicious"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:funny:because_its_true"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:practices_relating_to_the_transmission_of_genetic_information"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:kjhealy"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1301.3840">
    <title>[1301.3840] Utilities as Random Variables: Density Estimation and Structure Discovery</title>
    <dc:date>2013-01-19T14:43:43+00:00</dc:date>
    <link>http://arxiv.org/abs/1301.3840</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Decision theory does not traditionally include uncertainty over utility functions. We argue that the a person's utility value for a given outcome can be treated as we treat other domain attributes: as a random variable with a density function over its possible values. We show that we can apply statistical density estimation techniques to learn such a density function from a database of partially elicited utility functions. In particular, we define a Bayesian learning framework for this problem, assuming the distribution over utilities is a mixture of Gaussians, where the mixture components represent statistically coherent subpopulations. We can also extend our techniques to the problem of discovering generalized additivity structure in the utility functions in the population. We define a Bayesian model selection criterion for utility function structure and a search procedure over structures. The factorization of the utilities in the learned model, and the generalization obtained from density estimation, allows us to provide robust estimates of utilities using a significantly smaller number of utility elicitation questions. We experiment with our technique on synthetic utility data and on a real database of utility functions in the domain of prenatal diagnosis."

--- I thought about this problem a bit in 2010 (in the context of cost-benefit analysis), but am glad I didn't do anything with it, since this paper is from 2000...]]></description>
<dc:subject>to:NB decision_theory statistics estimation koller.daphne</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:16ba9777a087/</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:statistics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:estimation"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:koller.daphne"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.aeaweb.org/articles.php?doi=10.1257/jep.26.4.43">
    <title>Contingent Valuation: From Dubious to Hopeless</title>
    <dc:date>2012-11-06T17:33:20+00:00</dc:date>
    <link>http://www.aeaweb.org/articles.php?doi=10.1257/jep.26.4.43</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Approximately 20 years ago, Peter Diamond and I wrote an article for this journal analyzing contingent valuation methods. At that time Peter's view was that contingent valuation was hopeless, while I was dubious but somewhat more optimistic. But 20 years later, after millions of dollars of largely government-funded research, I have concluded that Peter's earlier position was correct and that contingent valuation is hopeless. In this paper, I selectively review the contingent valuation literature, focusing on empirical results. I find that three long-standing problems continue to exist: 1) hypothetical response bias that leads contingent valuation to overstatements of value; 2) large differences between willingness to pay and willingness to accept; and 3) the embedding problem which encompasses scope problems. The problems of embedding and scope are likely to be the most intractable. Indeed, I believe that respondents to contingent valuation surveys are often not responding out of stable or well-defined preferences, but are essentially inventing their answers on the fly, in a way which makes the resulting data useless for serious analysis. Finally, I offer a case study of a prominent contingent valuation study done by recognized experts in this approach, a study that should be only minimally affected by these concerns but in which the answers of respondents to the survey are implausible and inconsistent."]]></description>
<dc:subject>economics decision_theory cost-benefit_analysis in_NB</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1eac574df7d0/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cost-benefit_analysis"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1208.3213">
    <title>&quot;Ergodicity, Decisions, and Partial Information&quot; (Ramon van Handel)</title>
    <dc:date>2012-07-25T17:32:16+00:00</dc:date>
    <link>http://arxiv.org/abs/1208.3213</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["The aim of this paper is to exhibit connections between pathwise optimal strategies [for sequential prediction with partial information] and notions from ergodic theory. The sequential decision problem is developed in the general setting of an ergodic dynamical system ... with partial information ... . The existence of pathwise optimal strategies grounded in two basic properties: the conditional ergodic theory of the dynamical system, and the complexity of the loss function. When the loss function is not too complex, a general suﬃcient condition for the existence of pathwise optimal strategies is that the dynamical system is a conditional K-automorphism relative to the past observations ... If the conditional ergodicity assumption is strengthened, the complexity assumption can be weakened. Several examples demonstrate the interplay between complexity and ergodicity, which does not arise in the case of full information. Our results also yield a decision-theoretic characterization of weak mixing in ergodic theory, and establish pathwise optimality of ergodic nonlinear ﬁlters."

-Does this relate to when/whether we can have a uniform AEP?]]></description>
<dc:subject>ergodic_theory decision_theory information_theory van_handel.ramon stochastic_processes via:mraginsky have_read in_NB</dc:subject>
<dc:identifier>https://pinboard.in/u:cshalizi/b:1348f0a57d21/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:ergodic_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:information_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:van_handel.ramon"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:stochastic_processes"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:via:mraginsky"/>
	<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://www-stat.stanford.edu/~cgates/PERSI/papers/thinking.pdf">
    <title>The Problem of Thinking Too Much (Diaconis)</title>
    <dc:date>2012-07-19T22:32:00+00:00</dc:date>
    <link>http://www-stat.stanford.edu/~cgates/PERSI/papers/thinking.pdf</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA[Persi Diaconis: "Some years ago I was trying to decide whether or not to move to Harvard from Stanford. I had bored my friends silly with endless discussion. Finally, one of them said, 'You’re one of our leading decision theorists. Maybe you should make a list of the costs and benefits and try to roughly calculate your expected utility'. Without thinking, I blurted out, 'Come on, Sandy, this is serious.' "

--- I first heard this story with Jon Elster as the decision theorist.]]></description>
<dc:subject>decision_theory funny:academic funny:because_its_true diaconis.persi</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:48cecf01eaa2/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:funny:academic"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:funny:because_its_true"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:diaconis.persi"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://www.aeaweb.org/articles.php?doi=10.1257/jel.50.2.368">
    <title>Economic Incentives and Social Preferences: Substitutes or Complements?</title>
    <dc:date>2012-06-20T17:33:38+00:00</dc:date>
    <link>http://www.aeaweb.org/articles.php?doi=10.1257/jel.50.2.368</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["Explicit economic incentives designed to increase contributions to public goods and to promote other pro-social behavior sometimes are counterproductive or less effective than would be predicted among entirely self-interested individuals. This may occur when incentives adversely affect individuals' altruism, ethical norms, intrinsic motives to serve the public, and other social preferences. The opposite also occurs—crowding in—though it appears less commonly. In the fifty experiments that we survey, these effects are common, so that incentives and social preferences may be either substitutes (crowding out) or complements (crowding in). We provide evidence for four mechanisms that may account for these incentive effects on preferences: namely that incentives may (i) provide information about the person who implemented the incentive, (ii) frame the decision situation so as to suggest appropriate behavior, (iii) compromise a control averse individual's sense of autonomy, and (iv) affect the process by which people learn new preferences. An implication is that the evaluation of public policy must be restricted to allocations that are supportable as Nash equilibria when account is taken of these crowding effects. We show that well designed fines, subsidies, and the like minimize crowding out and may even do the opposite, making incentives and social preferences complements rather than substitutes."]]></description>
<dc:subject>to:NB economics institutions bowles.samuel experimental_economics mechanism_design decision_theory discipline_and_punish</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:e8030afa0758/</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:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:institutions"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bowles.samuel"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:experimental_economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:mechanism_design"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:discipline_and_punish"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://infostructuralist.wordpress.com/2012/06/01/information-theory-in-economics-part-i-rational-inattention/">
    <title>Information theory in economics, Part I: Rational inattention « The Information Structuralist</title>
    <dc:date>2012-06-02T00:10:51+00:00</dc:date>
    <link>http://infostructuralist.wordpress.com/2012/06/01/information-theory-in-economics-part-i-rational-inattention/</link>
    <dc:creator>cshalizi</dc:creator><dc:subject>information_theory decision_theory optimization economics raginsky.maxim</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:0bf2bbb32fe5/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:information_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:economics"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:raginsky.maxim"/>
</rdf:Bag></taxo:topics>
</item>
<item rdf:about="http://arxiv.org/abs/1202.3079">
    <title>[1202.3079] Towards minimax policies for online linear optimization with bandit feedback</title>
    <dc:date>2012-02-15T13:24:07+00:00</dc:date>
    <link>http://arxiv.org/abs/1202.3079</link>
    <dc:creator>cshalizi</dc:creator><description><![CDATA["We address the online linear optimization problem with bandit feedback. Our contribution is twofold. First, we provide an algorithm (based on exponential weights) with a regret of order $sqrt{d n log N}$ for any finite action set with $N$ actions, under the assumption that the instantaneous loss is bounded by 1. This shaves off an extraneous $sqrt{d}$ factor compared to previous works, and gives a regret bound of order $d sqrt{n log n}$ for any compact set of actions. Without further assumptions on the action set, this last bound is minimax optimal up to a logarithmic factor. Interestingly, our result also shows that the minimax regret for bandit linear optimization with expert advice in $d$ dimension is the same as for the basic $d$-armed bandit with expert advice. Our second contribution is to show how to use the Mirror Descent algorithm to obtain computationally efficient strategies with minimax optimal regret bounds in specific examples. More precisely we study two canonical action sets: the hypercube and the Euclidean ball. In the former case, we obtain the first computationally efficient algorithm with a $d sqrt{n}$ regret, thus improving by a factor $sqrt{d log n}$ over the best known result for a computationally efficient algorithm. In the latter case, our approach gives the first algorithm with a $sqrt{d n log n}$ regret, again shaving off an extraneous $sqrt{d}$ compared to previous works."]]></description>
<dc:subject>online_learning decision_theory optimization re:growing_ensemble_project cesa-bianchi.nicolo kakade.sham bubeck.sebastien in_NB bandit_problems</dc:subject>
<dc:source>https://pinboard.in/</dc:source>
<dc:identifier>https://pinboard.in/u:cshalizi/b:d3172d33e293/</dc:identifier>
<taxo:topics><rdf:Bag>	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:online_learning"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:decision_theory"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:optimization"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:re:growing_ensemble_project"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:cesa-bianchi.nicolo"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:kakade.sham"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bubeck.sebastien"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:in_NB"/>
	<rdf:li rdf:resource="https://pinboard.in/u:cshalizi/t:bandit_problems"/>
</rdf:Bag></taxo:topics>
</item>
</rdf:RDF>