BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Goals vs Utility Functions - Adrià Garriga Alonso (University of 
 Cambridge)
DTSTART:20190227T170000Z
DTEND:20190227T190000Z
UID:TALK120850@talks.cam.ac.uk
CONTACT:Adrià Garriga Alonso
DESCRIPTION:This week we read the “Goals vs Utility Functions” series 
 by Rohin Shah. This is the posts under “Ambitious Value Learning” at h
 ttps://www.lesswrong.com/s/4dHMdK5TLN6xcqtyc . You should read them before
  the session.\n\nOptional reading: "“Coherent Behaviour in the real worl
 d is an incoherent concept”":https://www.lesswrong.com/posts/vphFJzK3mWA
 4PJKAg/coherent-behaviour-in-the-real-world-is-an-incoherent by Richard Ng
 o.\n\nThese writings focus on the arguments that usually justify the premi
 se that a general AI will necessarily optimize a long-term\, explicit\, si
 mple goal. The authors find them to be insufficient\, and in the end propo
 se that perhaps a better approach to AGI safety is to construct agents wit
 hout long-term goals of this kind.\n\nAs usual\, there will be free pizza.
  The first half hour is for stragglers to finish reading.\n\nInvite your f
 riends to join the mailing list (https://lists.cam.ac.uk/mailman/listinfo/
 eng-safe-ai)\, the Facebook group (https://www.facebook.com/groups/1070763
 633063871) or the talks.cam page (https://talks.cam.ac.uk/show/index/80932
 ). Details about the next meeting\, the week’s topic and other events wi
 ll be advertised in these places.
LOCATION:Engineering Department\, CBL Seminar room BE4-38
END:VEVENT
END:VCALENDAR