Friday, May 07, 2021

All times are in EDT (UTC -4)

08:55-09:00 Introduction and Opening Remarks
09:00-09:30 Invited Talk Theorem Proving and Artificial Intelligence - A Brief Introduction.
Josef Urban, Czech Technical University in Prague
09:30-10:00 Invited Talk Reasoning with Deep Learning Architectures Based on System 2 Inductive Biases.
Yoshua Bengio, Mila, Université de Montréal
10:00-10:30 Poster Session 1
10:30-11:00 Coffee Break
11:00-11:30 Invited Talk The Relevance of Computational Creativity to Mathematical Reasoning Machines.
Alison Pease, University of Dundee
11:30-12:00 Invited Talk Towards A Human-Like Reasoning System.
Mateja Jamnik, University of Cambridge
12:00-12:30 Invited Talk Mathematical Reasoning in Humans.
Jay McClelland, Stanford University
12:30-13:00 Coffee Break
13:00-14:00 Discussion Panel Mathematical Reasoning for AGI: Challenge and Implications.
Yoshua Bengio, Timothy Gowers, Mateja Jamnik, Jay McClelland, Alison Pease, Christian Szegedy, Josef Urban.
14:00-14:30 Coffee Break
14:30-15:00 Invited Talk From Hammer to Scalpel: Progress in Automated Theorem Proving.
Markus Rabe, Google
15:00-15:30 Poster Session2
15:30-16:00 Coffee Break
16:00-16:30 Invited Talk Formal vs Informal Mathematics, Reasoning and AGI.
Stanislas Polu, OpenAI
16:30-16:50 Contributed Talk Training a First-Order Theorem Prover from Synthetic Data.
Vlad Firoiu
16:50-17:10 Contributed Talk Proof Artifact Co-training for Theorem Proving With Language Models.
Jesse Michael Han
17:10-17:30 Contributed Talk Improving Exploration in Policy Gradient Search: Application to Symbolic Optimization.
Mikel Landajuela Larma
17:30-17:35 Closing Remarks