BEGIN:VCALENDAR VERSION:2.0 PRODID:researchseminars.org CALSCALE:GREGORIAN X-WR-CALNAME:researchseminars.org BEGIN:VEVENT SUMMARY:Vladimir Vovk (University of London\, U.K.) DTSTART:20210319T093000Z DTEND:20210319T101500Z DTSTAMP:20260421T173915Z UID:BPS/25 DESCRIPTION:Title: Con formal prediction\nby Vladimir Vovk (University of London\, U.K.) as p art of Bangalore Probability Seminar\n\n\nAbstract\nhttps://www.isibang.ac .in/~statmath/pcm2020/\n\nMartingales in the foundations of statistics\nDa te: March 15th\, 2021\nTime: 15:00-15:45 IST\n\n\nAbstract: Traditional me thods of testing statistical hypotheses have been developed for the batch setting\, in the terminology of machine learning: given a batch of data\, statisticians typically compute measures of disagreement\, such as p-value s or Bayes factors\, between a null hypothesis and the data. An alternativ e that is popular in machine learning is the online setting\, in which the items of data (observations ) keep arriving sequentially. In this introdu ctory lecture I will explain the role of martingales\, in the form of test martingales\, in online hypothesis testing and discuss their applications in the foundations of probability and statistics.\n\nMultiple hypothesis testing with e-values\nDate: March 15th\, 2021\nTime: 16:00-16:45 IST\n\nA bstract: It is interesting that test martingales do not trivialize in the case of only one observation. In fact\, they provide a useful alternative\ , sometimes called e-values\, to the standard statistical notion of p-valu es. The most important mathematical advantage of e-values over p-values is that the average of e-values is always an e-value. This property is valua ble in multiple hypothesis testing\, which will be the topic of this lectu re.\n\nConformal prediction\nDate: March 19th\, 2021\nTime: 15:00-15:45 IS T\n\nAbstract: Mainstream machine learning\, despite its recent successes\ , has a serious drawback: while its state-of-the-art algorithms often prod uce excellent predictions\, they do not provide measures of their accuracy and reliability that would be both practically useful and provably valid. On the other hand\, such measures are commonplace in statistics. Conforma l prediction adapts rank tests\, popular in nonparametric statistics\, to testing the IID assumption (the observations being independent and identic ally distributed)\, which is the standard assumption made in machine learn ing. This gives us practical measures\, provably valid under the IID assum ption\, of the accuracy and reliability of predictions produced by traditi onal and recent machine-learning algorithms. In this lecture I will give a brief review of conformal prediction.\nConformal hypothesis testing\nDate : March 19th\, 2021\nTime: 16:00-16:45 IST\n\nAbstract: An interesting app lication of conformal prediction is the existence of exchangeability marti ngales\, i.e.\, random processes that are test martingales under any excha ngeable probability measure. In particular\, they are martingales whenever the observations are IID. The topics of this last lecture in this series will be the construction of exchangeability martingales and their use for different kinds of change detection\, including detecting a point at which the IID assumption becomes violated and detecting concept shift.\n LOCATION:https://researchseminars.org/talk/BPS/25/ END:VEVENT END:VCALENDAR