Proof Artifact Co-training for Theorem Proving with Language Models

02/11/2021 ∙ by Jesse Michael Han, et al. ∙ 19

Labeled data for imitation learning of theorem proving in large libraries of formalized mathematics is scarce as such libraries require years of concentrated effort by human specialists to be built. This is particularly challenging when applying large Transformer language models to tactic prediction, because the scaling of performance with respect to model size is quickly disrupted in the data-scarce, easily-overfitted regime. We propose PACT (Proof Artifact Co-Training), a general methodology for extracting abundant self-supervised data from kernel-level proof terms for co-training alongside the usual tactic prediction objective. We apply this methodology to Lean, an interactive proof assistant which hosts some of the most sophisticated formalized mathematics to date. We instrument Lean with a neural theorem prover driven by a Transformer language model and show that PACT improves theorem proving success rate on a held-out suite of test theorems from 32% to 48%.

READ FULL TEXT
POST COMMENT

Comments

There are no comments yet.

Authors

page 23

Code Repositories

lean-proof-recording-public-old

Recording of tactic proofs in Lean 3 for machine learning


view repo

lean-step-public

Proof artifact co-training for Lean


view repo

lean_proof_recording

Proof recording for Lean 3


view repo
This week in AI

Get the week's most popular data science and artificial intelligence research sent straight to your inbox every Saturday.