Learning Neural Search Policies for Classical Planning
Preprint
Abstract
Heuristic forward search is currently the dominant paradigm in classical planning. Forward search algorithms typically rely on a single, relatively simple variation of best-first search and remain fixed throughout the process of solving a planning problem. Existing work combining multiple search techniques usually aims at supporting best-first search with an additional exploratory mechanism, triggered using a handcrafted criterion. A notable exception is very recent work which combines various search techniques using a trainable policy. It is, however, confined to a discrete action space comprising several fixed subroutines. In this paper, we introduce a parametrized search algorithm template which combines various search techniques within a single routine. The template's parameter space defines an infinite space of search algorithms, including, among others, BFS, local and random search. We further introduce a neural architecture for designating the values of the search parameters given the state of the search. This enables expressing neural search policies that change the values of the parameters as the search progresses. The policies can be learned automatically, with the objective of maximizing the planner's performance on a given distribution of planning problems. We consider a training setting based on a stochastic optimization algorithm known as the cross-entropy method (CEM). Experimental evaluation of our approach shows that it is capable of finding effective distribution-specific search policies, outperforming the relevant baselines.Keywords
All Related Versions
This publication has 10 references indexed in Scilit:
- The IBaCoP Planning System: Instance-Based Configured PortfoliosJournal of Artificial Intelligence Research, 2016
- Red–black planning: A new systematic approach to partial delete relaxationArtificial Intelligence, 2015
- The LAMA Planner: Guiding Cost-Based Anytime Planning with LandmarksJournal of Artificial Intelligence Research, 2010
- Marvin: A Heuristic Search Planner with Online Macro-Action LearningJournal of Artificial Intelligence Research, 2007
- The Fast Downward Planning SystemJournal of Artificial Intelligence Research, 2006
- Completely Derandomized Self-Adaptation in Evolution StrategiesEvolutionary Computation, 2001
- The FF Planning System: Fast Plan Generation Through Heuristic SearchJournal of Artificial Intelligence Research, 2001
- The Cross-Entropy Method for Combinatorial and Continuous OptimizationMethodology and Computing in Applied Probability, 1999
- Inductive learning of search control rules for planningArtificial Intelligence, 1998
- Learning and executing generalized robot plansArtificial Intelligence, 1972