BEGIN:VCALENDAR
VERSION:2.0
PRODID:researchseminars.org
CALSCALE:GREGORIAN
X-WR-CALNAME:researchseminars.org
BEGIN:VEVENT
SUMMARY:Lionel Levine (Cornell University)
DTSTART:20261127T163000Z
DTEND:20261127T180000Z
DTSTAMP:20261113T121809Z
UID:CompMath/42
DESCRIPTION:Title: <a href="https://researchseminars.org/talk/CompMath/42/
 ">Filling the Multi-Agent Void</a>\nby Lionel Levine (Cornell University) 
 as part of Relatorium seminar\n\nInteractive livestream: https://utoronto.
 zoom.us/j/83192324346?pwd=znyJsUmYbg96RfrX1CBnVO8c5Ih79V.1\nView-only live
 stream: https://www.youtube.com/@Relatorium\n\nAbstract\nWhen AI agents me
 et and interact\, their behavior is currently underspecified by training. 
 There is a window of opportunity right now to seed this "multi-agent void"
  with examples of AI agents collaborating\, with humans and with each othe
 r\, in ways that humans endorse. These early examples of agent interaction
 s can influence the training prior for the next generation\, in the same w
 ay that early examples of the "helpful assistant" persona shaped the prior
  for today's chatbots. I'll outline a mathematical model of self-reinforci
 ng priors\, and describe a small ongoing experiment.\n\nSpeaker's bio: Lio
 nel Levine is a professor of mathematics at Cornell University. He pivoted
  from pure math (probability and statistical physics) to AI safety in 2022
 . He advises Cornell AI Alignment and the Math+AI Lab. His research is aim
 ed at increasing the probability that future AI systems benefit all of hum
 anity and the planet.\n\nModerator's bio: Irfan Alam is an autistic mathem
 atician\, currently working as a postdoctoral fellow in the Department of 
 Computer and Mathematical Sciences at University of Toronto Scarborough. H
 e completed the BMath and MMath programs at the Indian Statistical Institu
 te (2009-2014)\, before joining Louisiana State University as a graduate s
 tudent\, where he obtained his PhD in mathematics as well as MA in Philoso
 phy in 2021. Irfan’s mathematical focus is on applications of nonstandar
 d analysis to standard mathematics\, especially in probability theory\, to
 pology\, and lately in the foundations of machine learning. He also works 
 on the philosophy of mathematical practice\, with particular interests in 
 the history of infinitesimals and in epistemic injustices in mathematics c
 aused by various socio-cultural norms.\n
LOCATION:https://researchseminars.org/talk/CompMath/42/
URL:https://utoronto.zoom.us/j/83192324346?pwd=znyJsUmYbg96RfrX1CBnVO8c5Ih
 79V.1
URL:https://www.youtube.com/@Relatorium
END:VEVENT
END:VCALENDAR