When: Thursday, April 21, 2022, 3pm (CEST).

Where: Zoom online, check the address in the Google Calendar Event.

Topic: Target Languages (vs. Inductive Biases) for Learning to Act and Plan

Speaker: Prof. Hector Geffner, researcher at the Institució Catalana de Recerca i Estudis Avançats (ICREA) and professor at the Departament de Tecnologies de la Informació i les Comunicacions (DTIC), Universitat Pompeu Fabra, in the Artificial Intelligence Group. He is also a Wallenberg Guest Professor at Linköping University, Sweden.

Abstract

Recent breakthroughs in AI have shown the remarkable power of deep learning and deep reinforcement learning. These developments, however, have been tied to specific tasks, and progress in out-of-distribution generalization has been limited. While it is assumed that these limitations can be overcome by incorporating suitable inductive biases, the notion of inductive biases itself is often left vague and does not provide meaningful guidance. In the paper, I articulate a different learning approach where representations do not emerge from biases in a neural architecture but are learned over a given target language with a known semantics. The basic ideas are implicit in mainstream AI where representations have been encoded in languages ranging from fragments of first-order logic to probabilistic structural causal models. The challenge is to learn from data the representations that have traditionally been crafted by hand. Generalization is then a result of the semantics of the language. The goals of this paper are to make these ideas explicit, to place them in a broader context where the design of the target language is crucial, and to illustrate them in the context of learning to act and plan. For this, after a general discussion, I consider learning representations of actions, general policies, and subgoals (“intrinsic rewards”). In these cases, learning is formulated as a combinatorial problem but nothing prevents the use of deep learning techniques instead. Indeed, learning representations over languages with a known semantics provides an account of what is to be learned, while learning representations with neural nets provides a complementary account of how representations can be learned. The challenge and the opportunity is to bring the two together.

Short Bio

Hector Geffner is a researcher at the Institució Catalana de Recerca i Estudis Avançats (ICREA) and a professor at the Departament de Tecnologies de la Informació i les Comunicacions (DTIC), Universitat Pompeu Fabra, in the Artificial Intelligence Group. He is also a Wallenberg Guest Professor at Linköping University, Sweden.

Hector Geffner got his Ph.D at UCLA with a dissertation that was co-winner of the 1990 ACM Dissertation Award. He then worked as Staff Research Member at the IBM T.J. Watson Research Center in NY, USA and at the Universidad Simon Bolivar, in Caracas, Venezuela. Since 2001, he is a researcher at ICREA and a professor at the Universitat Pompeu Fabra, Barcelona, and since 2019, a Wallenberg Guest Professor at Linköping University, Sweden. Hector is a former Associate Editor of Artificial Intelligence and the Journal of Artificial Intelligence Research. He was also a member of the EurAI board, anjd is a Fellow of AAAI and EurAI. He is the author of the book ``Default Reasoning: Causal and Conditional Theories’’ , MIT Press, 1992, editor of “Heuristics, Probability, and Causality: a Tribute to Judea Pearl” along with R. Dechter and Joe Halpern, College Publications, 2010, and author of “A Concise Introduction to Models and Methods for Automated Planning” with Blai Bonet, Morgan and Claypool, 2013. Hector is interested in computational models of reasoning, action, learning, and planning that are general and effective. He is also a concerned citizen (particularly concerned these days) and, aside from courses on logic and AI, he teaches a course on social and technological change at the UPF. He leads a project on representation learning for planning (RLeap), funded by an Advanced ERC grant.

Material