Conflict History based Search for Constraint Satisfaction Problem

Djamal Habet (1) , Cyril Terrioux (1)

1 COALA - COntraintes, ALgorithmes et Applications

Branching heuristic is an important module in algorithms dedicated to solve Constraint Satisfaction Problems (CSP). It impacts the efficiency of exploring the search space and exploiting the problem structure. In this paper, we propose Conflict-History Search (CHS), a dynamic and adaptive branching heuristic for CSP solving. It is based on the search failures and considers the temporality of these failures throughout the resolution process. The exponential recency weighted average is used to estimate the evolution of the hardness of constraints throughout the search. The experimental evaluation on XCSP3 instances shows that integrating CHS to solvers based on MAC obtains competitive results and can improve those obtained by other heuristics of the state of the art, such as dom/wdeg and ABS.

Keywords: CSP; Branching Heuristic; Conflict Based Search; Search History; Exponential Recency Weighted

Domains: Computer Science [cs]; Artificial Intelligence [cs.AI]

Djamal Habet, Cyril Terrioux. Conflict History based Search for Constraint Satisfaction Problem. Proceedings of the 34th ACM/SIGAPP Symposium On Applied Computing (SAC), Apr 2019, Limassol, Cyprus. ⟨10.1145/3297280.3297389⟩. ⟨hal-02090618⟩