Towards Finding Longer Proofs

Zsolt Zombori, Adrián Csiszárik, Henryk Michalewski, Cezary Kaliszyk, Josef Urban

Keywords: automated reasoning, reinforcement learning, reasoning by analogy

Abstract: We present a reinforcement learning (RL) based guidance system for automated theorem proving geared towards Finding Longer Proofs (FLoP). FLoP is a step towards learning to reason by analogy, reducing the dependence on large scale search in automated theorem provers. We use several simple, structured datasets with very long proofs to show that FLoP can successfully generalise a single training proof to a large class of related problems, implementing a simple form of analogical reasoning. On these benchmarks, FLoP is competitive with strong theorem provers despite using very limited search.

One-sentence Summary: FLoP is a theorem prover that uses RL based guidance to implement a simple form of analogical reasoning to overcome fundamental limitations of search based approaches.

Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:1905.13100/code)

Reviewed Version (pdf): https://openreview.net/references/pdf?id=msZz1biPgh