Lecture: 'Generalized Planning With Graph Neural Networks'

We consider the problem of learning generalized policies for classical planning domains using graph neural networks (GNNs) from small instances represented in lifted STRIPS. The problem has been considered before but the proposed neural architectures are complex and the results are often mixed. In this work, we use a simple and general GNN architecture and aim at crisp experimental results and understanding: either the policy greedy in the learned value function achieves close to 100% generalization over larger instances than those in training, or the failure must be understood, and possibly fixed, logically. For this, we exploit the relation established between the expressive power of GNNs and the C2 fragment of first-order logic (namely, FOL with 2 variables and counting quantifiers).

The talk is aimed at a general audience with basic knowledge of computer science and artificial intelligence.

About

Blai Bonet is retired professor from the Computer Science Department at Universidad Simon Bolivar, Venezuela, and currently a research associate at the Universitat Pompeu Fabra, Barcelona, Spain. He received his Ph.D. in computer science from the University of California, Los Angeles. His research interests are in the areas of automated planning, search and knowledge representation, deep learning, and theory of computation. Blai has received several best paper awards or honorable mentions, including the 2009 and 2014 ICAPS Influential Paper Awards, and he is a co-author of the book "A Concise Introduction to Models and Methods for Automated Planning". He has served as associate editor of Artificial Intelligence and the Journal of Artificial Intelligence Research (JAIR), conference co-chair of ICAPS-12, program co-chair of AAAI-15, and has been a member of the Executive Council for ICAPS and AAAI.