BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Engineering Safe AI
SUMMARY:Machine Theory of Mind - Paolo Bova (University of
  Cambridge)
DTSTART;TZID=Europe/London:20190206T170000
DTEND;TZID=Europe/London:20190206T190000
UID:TALK119668AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/119668
DESCRIPTION:This week we will read about “Machine Theory of Mi
 nd”\, by Rabinowitz et al. who describe a neural n
 etwork\, their ToMnet\, which autonomously learns 
 how to model other agents' behaviour\, beliefs and
  intentions. ToMnet even recognises an important s
 eparation in the Theory of Mind literature between
  the beliefs that other agents hold about the envi
 ronment and its own. \n\nWhich features of their d
 esign allow ToMnet to infer the characteristics of
  other agents so successfully in their Gridworld e
 xperiments? And can we think of any challenges Tom
 Net would face in larger multi-agent environments?
 \n\nPaper: https://arxiv.org/abs/1802.07740\n\nWe 
 strongly recommend you read the paper before comin
 g. Regardless\, from 17:00 we will be reading the 
 paper and people can arrive. The discussion proper
  will start at 17:30. There will be pizza!
LOCATION:Cambridge University Engineering Department\, CBL 
 Seminar room BE4-38
CONTACT:Adrià Garriga Alonso
END:VEVENT
END:VCALENDAR