MAX-SAT problem using evolutionary algorithms

Abstract

MAX-SAT is a classic NP-hard optimization problem. Many real problems can be easily represented in, or reduced to MAX-SAT, and thus it has many applications. Finding optimum solutions of NP-hard optimization problems using limited computational resources seems infeasible in general. In particular, all known exact algorithms for MAX-SAT require worst-case exponential time, so evolutionary algorithms can be useful for finding good quality solutions in moderate time. We present the results of an experimental comparison of the performance of a number of recently proposed evolutionary algorithms for MAX-SAT. The algorithms include the Artificial Bee Colony (ABC) algorithm, Quantum Inspired Evolutionary Algorithm (QEA), Immune Quantum Evolutionary Algorithm (IQEA), Estimation of Distribution Algorithm (EDA), and randomized Monte Carlo (MC). Our experiments demonstrate that the ABC algorithm has better performance than the others. For problems with Boolean domain, such as MAX-SAT, the ABC algorithm requires specification of a suitable similarity measure. We experimentally evaluate the performance of the ABC algorithm with five different similarity measures to indicate the better choice for MAX-SAT problems.