Scalable Knowledge Refactoring using Constrained Optimisation

TL;DR

This paper presents a scalable knowledge refactoring method using constrained optimization, encoding decisions at the literal level and focusing on linear rules, achieving faster and more compressed results than previous methods.

Contribution

Introduces a novel constrained optimization approach for knowledge refactoring that scales to large programs by encoding decisions with literals and emphasizing linear rules.

Findings

Refactors programs faster than previous methods

Achieves up to 60% more compression

Effective on multiple domains

Abstract

Knowledge refactoring compresses a logic program by introducing new rules. Current approaches struggle to scale to large programs. To overcome this limitation, we introduce a constrained optimisation refactoring approach. Our first key idea is to encode the problem with decision variables based on literals rather than rules. Our second key idea is to focus on linear invented rules. Our empirical results on multiple domains show that our approach can refactor programs quicker and with more compression than the previous state-of-the-art approach, sometimes by 60%.