Smarter Features, Simpler Learning?

TL;DR

This paper explores the use of complex, structural features in machine learning for automated reasoning, aiming to improve strategy learning and tool selection in theorem proving and term rewriting.

Contribution

It introduces the concept of using structural features from the software verification community to enhance machine learning models in automated reasoning tasks.

Findings

Proposes new structural features for term rewrite systems and theorem proving.

Suggests potential for improved tool selection and strategy learning.

Lays groundwork for future empirical evaluation.

Abstract

Earlier work on machine learning for automated reasoning mostly relied on simple, syntactic features combined with sophisticated learning techniques. Using ideas adopted in the software verification community, we propose the investigation of more complex, structural features to learn from. These may be exploited to either learn beneficial strategies for tools, or build a portfolio solver that chooses the most suitable tool for a given problem. We present some ideas for features of term rewrite systems and theorem proving problems.