Holophrasm: a neural Automated Theorem Prover for higher-order logic
This addresses the problem of automating complex theorem proving for researchers in formal verification and AI, though it appears incremental as it builds on existing neural methods for a specific logic.
The paper tackles automated theorem proving in higher-order logic by proposing a system that uses deep learning without hand-constructed features, achieving a result of proving 14% of test theorems from Metamath's set.mm module.
I propose a system for Automated Theorem Proving in higher order logic using deep learning and eschewing hand-constructed features. Holophrasm exploits the formalism of the Metamath language and explores partial proof trees using a neural-network-augmented bandit algorithm and a sequence-to-sequence model for action enumeration. The system proves 14% of its test theorems from Metamath's set.mm module.