USING ARTIFICIAL NEURAL NETWORKS FOR CONSTRAINT SATISFACTION PROBLEM

We address the problem of solving a constraint satisfaction problem (C SP) by treating a constraint logic program (CLP) as a network of const raints. We attempt to show that each computation in a CLP becomes a se quence of linear steps, since the check satisfiability of the system o f constraints is applied at each resolution step which is linear in th e size of the current constraint problem. Thus, the constraint propaga tion information is performed at each step during any CLP derivation. The major issues we address here are the identification (using logic i nterpretation) of constraints that can be added within the program rul es to reduce the size of intermediate states and how to use the previo us steps of the computation as a guidance for CLP derivations.