Bayesian Optimization Algorithm for SAT Problem Incorporating Heuristic Operator
Yan Xiang Wu · 2008
Univariate marginal distribution algorithm(UMDA)is a new evolutionary algorithm and is an efficient algo- rithm for complicate problem.According to the characteristics of SAT problem,an improved UMDA for solving SAT problem(HeUMDASAT)is proposed in this paper.A new heuristic operator is presented by combing the structure in- formation of SAT problem with the global information extracted from promising individuals of current population.Dif- ferent from the stochastic searching operator,the heuristic operator can generate new promising individuals and help the UMDA skip from the local optimum.In the experiments,3700 benchmark SAT problems in SATL IB are used to test the performance of HeUMDASAT.The experiments results show that the HeUMDASAT makes progress in the aspects of speed and the success rate.