Inference and Learning in Planning
(invited lecture for DS 2009)
Author: Hector Geffner
Affiliation:
ICREA & Universitat Pompeu Fabra, Barcelona, Spain
Abstract.
Planning is concerned with the development of solvers for a wide range
of models where actions must be selected for achieving goals. In these
models, actions may be deterministic or not, full or partial sensing
may be available or not, and so on. In the last few years, significant
progress has been made, resulting in algorithms that can produce plans
effectively in a variety of settings. These developments have to do
with the formulation and use of general inference techniques and
transformations.
In this talk, I'll review the inference techniques used for solving
individual planning instances, and discuss the use of learning methods
and transformations for solving complete planning domains. The former
methods lead to the automatic derivations of heuristic functions; the
latter, to the automatic derivation of compact policies and useful
domain concepts.
Biography:
Hector Geffner got his Ph.D in UCLA with a dissertation that was
co-winner of the 1990 ACM Dissertation Award. Then he 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. He is
currently a researcher at ICREA and a professor at the Universitat
Pompeu Fabra in Barcelona, Spain. His work in the last ten years has
been in the area of planning. Hector Geffner is a AAAI fellow,
Associate Editor of JAIR, and Editorial Board member at AIJ.
©Copyright 2009 Author
|