BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Drupal iCal API//EN
X-WR-CALNAME:Events items teaser
X-WR-TIMEZONE:America/Toronto
BEGIN:VTIMEZONE
TZID:America/Toronto
X-LIC-LOCATION:America/Toronto
BEGIN:DAYLIGHT
TZNAME:EDT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
DTSTART:20230312T070000
END:DAYLIGHT
BEGIN:STANDARD
TZNAME:EST
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
DTSTART:20221106T060000
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:656dd0d224394
DTSTART;TZID=America/Toronto:20230726T120000
SEQUENCE:0
TRANSP:TRANSPARENT
DTEND;TZID=America/Toronto:20230726T130000
SUMMARY:IQC Student Seminar Featuring Evan Peters
CLASS:PUBLIC
DESCRIPTION:Summary \n\nSOME LEARNING BOUNDS AND GUARANTEES FOR TESTING (QU
 ANTUM) HYPOTHESES\n\nMachine learning is a powerful tool\, yet we often do
  not know how well\na learning algorithm might perform on any given task. 
 One standard\napproach to bound the accuracy of a learning algorithm is to
  reduce\nthe learning task to hypothesis testing. Fano's inequality then s
 tates\nthat a large amount of mutual information between the learner's\nob
 servations and the set of unknown parameters is a necessary\ncondition for
  success. \n\nIn this talk\, I will describe how such a condition is also 
 sufficient\nfor succeeding at some learning task\, thereby providing a pur
 ely\ninformation-theoretic guarantee for learning. Noting that this\nguara
 ntee has an immediate extension to quantum information theory\, I\nwill th
 en introduce the task of \"testing quantum hypotheses\"\, in which\nthe un
 known parameters of the learning task are prepared in a quantum\nregister 
 in superposition (rather than being sampled stochastically)\nand the learn
 er's success at this task is measured by their ability to\nestablish quant
 um correlations with that register. I will discuss\nongoing attempts to ch
 aracterize this scenario.\n\n \n\nADD EVENT TO CALENDAR\n\nApple [https:/
 /www.addevent.com/event/FO18051378+apple]  Google\n[https://www.addevent.
 com/event/FO18051378+google]  Office 365\n[https://www.addevent.com/event
 /FO18051378+office365]  Outlook\n[https://www.addevent.com/event/FO180513
 78+outlook]  Outlook.com\n[https://www.addevent.com/event/FO18051378+outl
 ookcom]  Yahoo\n[https://www.addevent.com/event/FO18051378+yahoo]  \n 
DTSTAMP:20231204T131458Z
END:VEVENT
END:VCALENDAR