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:20210319T103000Z
DTEND:20210319T111500Z
DTSTAMP:20260421T173821Z
UID:BPS/26
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/BPS/26/">Con
 formal hypothesis testing</a>\nby Vladimir Vovk (University of London\, U.
 K.) as part of Bangalore Probability Seminar\n\n\nAbstract\nhttps://www.is
 ibang.ac.in/~statmath/pcm2020/\n\nMartingales in the foundations of statis
 tics\nDate: March 15th\, 2021\nTime: 15:00-15:45 IST\n\n\nAbstract: Tradit
 ional methods of testing statistical hypotheses have been developed for th
 e batch setting\, in the terminology of machine learning: given a batch of
  data\, statisticians typically compute measures of disagreement\, such as
  p-values or Bayes factors\, between a null hypothesis and the data. An al
 ternative that is popular in machine learning is the online setting\, in w
 hich the items of data (observations ) keep arriving sequentially. In this
  introductory lecture I will explain the role of martingales\, in the form
  of test martingales\, in online hypothesis testing and discuss their appl
 ications in the foundations of probability and statistics.\n\nMultiple hyp
 othesis testing with e-values\nDate: March 15th\, 2021\nTime: 16:00-16:45 
 IST\n\nAbstract: It is interesting that test martingales do not trivialize
  in the case of only one observation. In fact\, they provide a useful alte
 rnative\, sometimes called e-values\, to the standard statistical notion o
 f p-values. The most important mathematical advantage of e-values over p-v
 alues is that the average of e-values is always an e-value. This property 
 is valuable in multiple hypothesis testing\, which will be the topic of th
 is lecture.\n\nConformal prediction\nDate: March 19th\, 2021\nTime: 15:00-
 15:45 IST\n\nAbstract: Mainstream machine learning\, despite its recent su
 ccesses\, has a serious drawback: while its state-of-the-art algorithms of
 ten produce excellent predictions\, they do not provide measures of their 
 accuracy and reliability that would be both practically useful and provabl
 y valid. On the other hand\, such measures are commonplace in statistics. 
 Conformal prediction adapts rank tests\, popular in nonparametric statisti
 cs\, to testing the IID assumption (the observations being independent and
  identically distributed)\, which is the standard assumption made in machi
 ne learning. This gives us practical measures\, provably valid under the I
 ID assumption\, of the accuracy and reliability of predictions produced by
  traditional and recent machine-learning algorithms. In this lecture I wil
 l give a brief review of conformal prediction.\nConformal hypothesis testi
 ng\nDate: March 19th\, 2021\nTime: 16:00-16:45 IST\n\nAbstract: An interes
 ting application of conformal prediction is the existence of exchangeabili
 ty martingales\, i.e.\, random processes that are test martingales under a
 ny exchangeable 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/26/
END:VEVENT
END:VCALENDAR