Passer à la navigation principale Passer à la recherche Passer au contenu principal

HyperTree Proof Search for Neural Theorem Proving

  • Guillaume Lample
  • , Marie Anne Lachaux
  • , Thibaut Lavril
  • , Xavier Martinet
  • , Amaury Hayat
  • , Gabriel Ebner
  • , Aurélien Rodriguez
  • , Timothée Lacroix
  • Meta AI
  • Vrije Universiteit Amsterdam

Résultats de recherche: Le chapitre dans un livre, un rapport, une anthologie ou une collectionContribution à une conférenceRevue par des pairs

Résumé

We propose an online training procedure for a transformer-based automated theorem prover. Our approach leverages a new search algorithm, HyperTree Proof Search (HTPS), that learns from previous proof searches through online training, allowing it to generalize to domains far from the training distribution. We report detailed ablations of our pipeline's main components by studying performance on three environments of increasing complexity. In particular, we show that with HTPS alone, a model trained on annotated proofs manages to prove 65.4% of a held-out set of Metamath theorems, significantly outperforming the previous state of the art of 56.5% by GPT-f. Online training on these unproved theorems increases accuracy to 82.6%. With a similar computational budget, we improve the state of the art on the Lean-based miniF2F-curriculum dataset from 31% to 42% proving accuracy.

langue originaleAnglais
titreAdvances in Neural Information Processing Systems 35 - 36th Conference on Neural Information Processing Systems, NeurIPS 2022
rédacteurs en chefS. Koyejo, S. Mohamed, A. Agarwal, D. Belgrave, K. Cho, A. Oh
EditeurNeural information processing systems foundation
ISBN (Electronique)9781713871088
étatPublié - 1 janv. 2022
Evénement36th Conference on Neural Information Processing Systems, NeurIPS 2022 - New Orleans, États-Unis
Durée: 28 nov. 20229 déc. 2022

Série de publications

NomAdvances in Neural Information Processing Systems
Volume35
ISSN (imprimé)1049-5258

Une conférence

Une conférence36th Conference on Neural Information Processing Systems, NeurIPS 2022
Pays/TerritoireÉtats-Unis
La villeNew Orleans
période28/11/229/12/22

Empreinte digitale

Examiner les sujets de recherche de « HyperTree Proof Search for Neural Theorem Proving ». Ensemble, ils forment une empreinte digitale unique.

Contient cette citation