Machine Learning in Proof General: Interfacing Interfaces

TL;DR

ML4PG is a machine learning extension for Proof General that analyzes proof structures and provides proof hints by clustering proof data using MATLAB and Weka.

Contribution

It introduces an automated interface for integrating machine learning with interactive theorem proving in Coq and SSReflect.

Findings

Proof data clustering improves proof guidance.

Automated interfacing streamlines proof analysis.

Enhanced proof development experience.

Abstract

We present ML4PG - a machine learning extension for Proof General. It allows users to gather proof statistics related to shapes of goals, sequences of applied tactics, and proof tree structures from the libraries of interactive higher-order proofs written in Coq and SSReflect. The gathered data is clustered using the state-of-the-art machine learning algorithms available in MATLAB and Weka. ML4PG provides automated interfacing between Proof General and MATLAB/Weka. The results of clustering are used by ML4PG to provide proof hints in the process of interactive proof development.