Combining inference and search produces successful schemes for solving constraint satisfaction problems. Based on this idea a general scheme which uses inference inside evolutionary computation techniques is presented. A genetic algorithm and the particle swarm optimization heuristic make use of adaptable inference levels offered by the mini-bucket elimination algorithm. Experimental results prove the efficiency of our approach in solving the Max-CSP optimization task. The inference/search trade-off is analyzed
This article is authored also by Synbrain data scientists and collaborators. READ THE FULL ARTICLE