{"98691":{"#nid":"98691","#data":{"type":"event","title":"ARC Colloquium: Ken Regan, University at Buffalo (SUNY)","body":[{"value":"\u003Cp\u003E\u003Cstrong\u003EAbstract:\u003C\/strong\u003E\u003C\/p\u003E\u003Cp\u003EWe consider the problem of inferring probabilistic behavior by agents faced with decision\u003Cbr \/\u003Eoptions m_1,m_2,...,m_n, in terms of hindsight utility values u_1,u_2,...,u_n and parameters Z\u003Cbr \/\u003Egoverning the aptitude of the agent.\u0026nbsp; In chess the options are the legal moves in a given\u003Cbr \/\u003Eposition, the utilities are values computed by strong chess programs, and the parameters\u003Cbr \/\u003Eare fitted to the international chess Elo rating scale.\u0026nbsp; We show with large data that Bayesian\u003Cbr \/\u003Eand maximum-likelihood methods are markedly inferior to simple frequentist methods at\u003Cbr \/\u003Ethis task.\u0026nbsp; We justify theoretically our contention that the former methods emphasize the \u003Cbr \/\u003Eoption that was actually chosen at each turn in the training sets in ways that fail to use\u003Cbr \/\u003Emuch of the information in the data.\u003C\/p\u003E\u003Cp\u003EThe statistical model was developed with Guy Haworth (Univ. of Reading, UK) in papers at \u003Cbr \/\u003EAAAI 2011 and the 2011 Advances in Computer Games conference.\u0026nbsp; The talk will also show\u003Cbr \/\u003Ehow it is employed to compute \u0022Intrinsic Ratings\u0022 based on quality of moves made rather than\u003Cbr \/\u003Ethe results of games, and to evaluate statistical allegations of players cheating with computer\u003Cbr \/\u003Eprograms during games.\u0026nbsp; Unlike many field studies of decision making the data sets have\u003Cbr \/\u003Ebeen taken under real competition, and the talk will discuss attendant issues of inference\u003Cbr \/\u003Efrom large data, handling it, caveats in interpreting it, and the general practice of science.\u003C\/p\u003E\u003Cp\u003E\u003Cbr \/\u003E\u0026nbsp;\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":"","field_summary_sentence":[{"value":"Bayes Meets Waterloo at Chess?"}],"uid":"27263","created_gmt":"2012-02-01 09:32:43","changed_gmt":"2016-10-08 01:57:48","author":"Elizabeth Ndongi","boilerplate_text":"","field_publication":"","field_article_url":"","field_event_time":{"event_time_start":"2012-02-07T12:00:00-05:00","event_time_end":"2012-02-07T12:00:00-05:00","event_time_end_last":"2012-02-07T12:00:00-05:00","gmt_time_start":"2012-02-07 17:00:00","gmt_time_end":"2012-02-07 17:00:00","gmt_time_end_last":"2012-02-07 17:00:00","rrule":null,"timezone":"America\/New_York"},"extras":[],"groups":[{"id":"50875","name":"School of Computer Science"},{"id":"70263","name":"ARC"}],"categories":[],"keywords":[],"core_research_areas":[],"news_room_topics":[],"event_categories":[{"id":"1795","name":"Seminar\/Lecture\/Colloquium"}],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Ca href=\u0022mailto:ndongi@cc.gatech.edu\u0022\u003Endongi@cc.gatech.edu\u003C\/a\u003E\u003C\/p\u003E","format":"limited_html"}],"email":[],"slides":[],"orientation":[],"userdata":""}}}