# Reinforcement Learning of Theorem Proving

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)

## Introduction

Automated reasoning over mathematical proof was a major motivation for the development of computer science. Automated theorem provers (ATP) can in principle be used to attack any formally stated mathematical problem, and is a research area that has been studied since the early 20th century.  As of today, state-of-art ATP systems rely on fast implementation of complete proof calculi. such as resolution and tableau. However, they are still far weaker than trained mathematicians. Within current ATP systems, many heuristics are essential for their performance. As a result, in recent years machine learning has been used to replace such heuristics and improve the performance of ATPs.

In this paper the authors implemented a reinforcement learning based ATP, rlCoP. The proposed ATP reasons within first-order logic. The underlying proof calculi is the connection calculi , and the reinforcement learning method is Monte Carlo tree search along with policy and value learning. It was shown that reinforcement learning results in a 42.1% performance increase compared to the base prover (without learning).

## Related Work

C. Kalizyk and J. Urban proposed a supervised learning based ATP, FEMaLeCoP, whose underlying proof calculi is the same as this paper in 2015.  In their algorithm learns from existing proofs to choose the next tableau extension step. Similarly, S. Loos, et al. developed an supervised learning ATP system in 2017 , with superposition as their proof calculi. However they chose deep neural network (CNNs and RNNs) as feature extractor.

On a different note, A. Alemi, et al. proposed a deep sequence model for premise selection in 2016 , and they claim to be the first team to involve deep neural networks in ATPs. Although premise selection is not directly linked to automated reasoning, it is still an important component in ATPs, and their paper provides some insights about how to process datasets of formally stated mathematical problems.

## First Order Logic and Connection Calculi

Here we assume basic first-order logic and theorem proving terminology, and we will offer a brief introduction of the base prover and connection calculi. For example, let us try to prove the following first-order sentence.