top of page


Proof Truth

Jared Able, Hongyi Shen, Dennis Nguyen, Evgeniya Lagoda, Zhihan Li


Can AI learn to do logic? We'll explore this question by teaching an AI to fill in gaps in mathematical proofs. After training a model on tens of thousands of mathematical proofs, we'll apply reinforcement learning methods like DQN to improve performance.

Preferable background (though not necessary if you're willing to learn!): experience with NLP, Markov chains, Monte Carlo simulations, or deep learning in general

Screen Shot 2022-06-03 at 11.31.35 AM.png
github URL
bottom of page