主催: The Japanese Society for Artificial Intelligence
会議名: 2017年度人工知能学会全国大会（第31回）
回次: 31
開催地: 愛知県名古屋市 ウインクあいち
開催日: 2017/05/23 - 2017/05/26
Most state-of-the-art SAT solvers use VSIDS as their branching heuristic because of its robustness. However, in general, a branching heuristic can not cover all problems. It is axiomatic that mixture of branching heuristic with adequate configuration performs better than single branching heuristic. In this paper, we propose hybrid branching heuristic and show some preliminary experimental results using benchmarks from SAT competitions to evaluate its efficiency.
