Maximizing over multiple pattern databases speeds up heuristic search

A pattern database (PDB) is a heuristic function stored as a lookup table. This paper considers how best to use a fixed amount (m units) of memory for storing pattern databases. In particular, we examine whether using n pattern databases of size m/n instead of one pattern database of size m improves search performance. In all the state spaces considered, the use of multiple smaller pattern databases reduces the number of nodes generated by IDA*. The paper provides an explanation for this phenomenon based on the distribution of heuristic values that occur during search.

[1]  Eric A. Hansen,et al.  Space-Efficient Memory-Based Heuristics , 2004, AAAI.

[2]  Robert C. Holte,et al.  Experiments with Automatically Created Memory-Based Heuristics , 2000, SARA.

[3]  Larry S. Davis,et al.  Pattern Databases , 1979, Data Base Design Techniques II.

[4]  Richard E. Korf,et al.  Depth-First Iterative-Deepening: An Optimal Admissible Tree Search , 1985, Artif. Intell..

[5]  Richard E. Korf,et al.  Best-First Frontier Search with Delayed Duplicate Detection , 2004, AAAI.

[6]  Robert C. Holte,et al.  Hierarchical A*: Searching Abstraction Hierarchies Efficiently , 1996, AAAI/IAAI, Vol. 1.

[7]  Stefan Edelkamp,et al.  Abstraction in directed model checking , 2004 .

[8]  Albert Nymeyer,et al.  Guided Invariant Model Checking Based on Abstraction and Symbolic Pattern Databases , 2004, TACAS.

[9]  Jonathan Schaeffer,et al.  Dual Search in Permutation State Spaces , 2006, AAAI.

[10]  Jonathan Schaeffer,et al.  Efficiently Searching the 15-Puzzle , 1994 .

[11]  Brian Tanner,et al.  Hierarchical Heuristic Search Revisited , 2005, SARA.

[12]  Eric A. Hansen,et al.  Structured Duplicate Detection in External-Memory Graph Search , 2004, AAAI.

[13]  Jonathan Schaeffer,et al.  Pattern Databases , 1998, Comput. Intell..

[14]  Giovanni Manzini,et al.  BIDA: An Improved Perimeter Search Algorithm , 1995, Artif. Intell..

[15]  István T. Hernádvölgyi Automatically generated lower bounds for search , 2004 .

[16]  Richard E. Korf,et al.  Additive Pattern Database Heuristics , 2004, J. Artif. Intell. Res..

[17]  Hermann Kaindl,et al.  A Case Study of Revisiting Best-First vs. Depth-First Search , 2004, ECAI.

[18]  Jonathan Schaeffer,et al.  Dual Lookups in Pattern Databases , 2005, IJCAI.

[19]  David Furcy,et al.  Multiple Pattern Databases , 2004, ICAPS.

[20]  Eric A. Hansen,et al.  Breadth-first heuristic search , 2004, Artif. Intell..

[21]  Richard E. Korf,et al.  Compressing Pattern Databases , 2004, AAAI.

[22]  Richard E. Korf,et al.  Finding Optimal Solutions to Rubik's Cube Using Pattern Databases , 1997, AAAI/IAAI.

[23]  István T. Hernádvölgyi Solving the Sequential Ordering Problem with Automatically Generated Lower Bounds , 2004 .

[24]  Richard E. Korf,et al.  Time complexity of iterative-deepening-A* , 2001, Artif. Intell..

[25]  Patrick Brézillon,et al.  Lecture Notes in Artificial Intelligence , 1999 .

[26]  Nils J. Nilsson,et al.  A Formal Basis for the Heuristic Determination of Minimum Cost Paths , 1968, IEEE Trans. Syst. Sci. Cybern..

[27]  Richard E. Korf,et al.  Disjoint pattern database heuristics , 2002, Artif. Intell..

[28]  Toby Walsh,et al.  Symposium on Abstraction, Reformulation, and Approximation (SARA-2000) , 2001, AI Mag..