An Empirical Evaluation of Portfolios Approaches for solving CSPs

TL;DR

This paper empirically evaluates portfolio approaches for solving CSPs, showing that SAT-based methods perform best and outperform simple classification models in solving efficiency.

Contribution

It compares different portfolio models for CSPs, adapting SAT approaches and machine learning, highlighting the superior performance of SAT-based portfolios.

Findings

SAT approaches outperform other models in CSP solving

Portfolio size impacts performance but SAT methods remain top

Simple classification models are less competitive

Abstract

Recent research in areas such as SAT solving and Integer Linear Programming has shown that the performances of a single arbitrarily efficient solver can be significantly outperformed by a portfolio of possibly slower on-average solvers. We report an empirical evaluation and comparison of portfolio approaches applied to Constraint Satisfaction Problems (CSPs). We compared models developed on top of off-the-shelf machine learning algorithms with respect to approaches used in the SAT field and adapted for CSPs, considering different portfolio sizes and using as evaluation metrics the number of solved problems and the time taken to solve them. Results indicate that the best SAT approaches have top performances also in the CSP field and are slightly more competitive than simple models built on top of classification algorithms.