BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Integrating Combinatorial Solvers and Neural Models - Pasquale Min
 ervi\, University of Edinburgh
DTSTART:20240223T120000Z
DTEND:20240223T130000Z
UID:TALK211501@talks.cam.ac.uk
CONTACT:Richard Diehl Martinez
DESCRIPTION:Neural models -- including language models such as ChatGPT -- 
 can exhibit remarkable abilities\; paradoxically\, they also struggle with
  algorithmic tasks where much simpler models excel. To solve these issues\
 , we propose Implicit Maximum Likelihood Estimation (IMLE)\, a framework f
 or end-to-end learning of models combining algorithmic combinatorial solve
 rs and differentiable neural components\, which allows us to incorporate p
 lanning and reasoning algorithms in neural architectures by just adding a 
 simple decorator [1\, 2].\n\n[1] Implicit MLE: Backpropagating Through Dis
 crete Exponential Family Distributions. https://arxiv.org/abs/2106.01798\,
  NeurIPS 2021\n[2] Adaptive Perturbation-Based Gradient Estimation for Dis
 crete Latent Variable Models. https://arxiv.org/abs/2209.04862\, AAAI 2023
 \n\nJoin Zoom Meeting\n\nhttps://cam-ac-uk.zoom.us/j/86071371348?pwd=OVlqd
 DhZNHlGbzV5RUZrSzM1cUlhUT09\n\n \n\nMeeting ID: 860 7137 1348\n\nPasscode:
  387918\n\n \n
LOCATION:https://cam-ac-uk.zoom.us/j/86071371348?pwd=OVlqdDhZNHlGbzV5RUZrS
 zM1cUlhUT09
END:VEVENT
END:VCALENDAR