Using machine learning to make constraint solver implementation decisions

TL;DR

This paper explores using machine learning to automate and optimize design decisions in constraint solver implementations, demonstrated with the alldifferent constraint, leading to improved solver performance.

Contribution

It introduces a machine learning approach for automatic, problem-dependent decision-making in constraint solver design, surpassing default choices.

Findings

Machine learning can effectively guide constraint solver decisions.

Automated decisions improve solver performance over default choices.

The system adapts to different problem instances for better results.

Abstract

Programs to solve so-called constraint problems are complex pieces of software which require many design decisions to be made more or less arbitrarily by the implementer. These decisions affect the performance of the finished solver significantly. Once a design decision has been made, it cannot easily be reversed, although a different decision may be more appropriate for a particular problem. We investigate using machine learning to make these decisions automatically depending on the problem to solve with the alldifferent constraint as an example. Our system is capable of making non-trivial, multi-level decisions that improve over always making a default choice.