The 8th Italian Workshop on Planning and Scheduling (IPS-2020) aims at bringing together researchers interested in different aspects of planning and scheduling problems, and to introduce new researchers to the community. Although the primary target of this series of workshops is the Italian community of P&S, the aim is to attract an international gathering, thus expecting contributions and participations from around the world.
IPS-2020 will be held online on November 25th in conjunction with the 19th International Conference of the Italian Association for Artificial Intelligence – AIxIA 2020 (reboot), Anywhere, November 25th-27th, 2020.
Invited Talk
Width, Serializations, and General Policies
Hector Geffner
ICREA and Universitat Pompeu Fabra
Abstract
There are three important notions in planning that appear to be critical, both practically and theoretically, but which have not been sufficiently studied. One is the notion of bounded width, a characteristic of many planning domains when goals are atomic, and which allows for polynomial search algorithms and extensions as found in state-of-the-art planners like Dual-BFWS. Second, the notion of serialization; namely, the idea that planning problems can be solved often by solving a sequence of subproblems. And third, the notion of general policies: policies that solve not just one problem instance but a whole collection of them. In this talk, I’ll present recent work with Blai that ties three notions together. An interesting side-effect of the analysis is a new powerful language for specifying serializations that goes well beyond current methods based on goal or landmark counters, or heuristics. The language of policy sketches can represent from basic serializations to full general policies.
About the speaker
Hector Geffner is an ICREA Research Professor at the Universitat Pompeu Fabra (UPF) in Barcelona, Spain, and a Wallenberg Guest Professor at Linköping University. He was born and grew up in Buenos Aires and obtained a PhD in Computer Science at UCLA in 1989 under the supervision of Judea Pearl. He worked then at the IBM T.J. Watson Research Center in NY, USA, and at the Universidad Simon Bolivar, in Caracas. Hector is a Fellow of AAAI and EurAI. He teaches logic, AI, and novel course on social and technological change. He has recently obtained an Advanced ERC grant to do research on symbolic representation learning for planning.
More recent editions:
Sponsor
