BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:AI Safety via Debate - Beth Barnes (University of Cambridge)
DTSTART:20180509T160000Z
DTEND:20180509T173000Z
UID:TALK105784@talks.cam.ac.uk
CONTACT:Adrià Garriga Alonso
DESCRIPTION:The seminar will be on OpenAI's new AI safety model\, https://
 blog.openai.com/debate/ .\n\nHow can we augment humans so that they can ef
 fectively supervise advanced AI systems? One way is to take advantage of t
 he AI itself to help with the supervision\, asking the AI (or a separate A
 I) to point out flaws in any proposed action. To achieve this\, we reframe
  the learning problem as a game played between two agents\, where the agen
 ts have an argument with each other and the human judges the exchange. Eve
 n if the agents have a more advanced understanding of the problem than the
  human\, the human may be able to judge which agent has the better argumen
 t (similar to expert witnesses arguing to convince a jury).\n\nBring a lap
 top if you can\, so we can play the cat/dog game! https://debate-game.open
 ai.com/\n
LOCATION: Cambridge University Engineering Department\, CBL Seminar room B
 E4-38.  For directions see http://learning.eng.cam.ac.uk/Public/Directions
END:VEVENT
END:VCALENDAR