TacticToe: Learning to Prove with Tactics
arXiv:1804.00596 · doi:10.1007/s10817-020-09580-x
Abstract
We implement a automated tactical prover TacticToe on top of the HOL4 interactive theorem prover. TacticToe learns from human proofs which mathematical technique is suitable in each proof situation. This knowledge is then used in a Monte Carlo tree search algorithm to explore promising tactic-level proof paths. On a single CPU, with a time limit of 60 seconds, TacticToe proves 66.4 percent of the 7164 theorems in HOL4's standard library, whereas E prover with auto-schedule solves 34.5 percent. The success rate rises to 69.0 percent by combining the results of TacticToe and E prover.
References in corpus (1)
Cited by in corpus (13)
- Learning Guided Automated Reasoning: A Brief Survey
- LIME: Learning Inductive Bias for Primitives of Mathematical Reasoning
- TacticZero: Learning to Prove Theorems from Scratch with Deep Reinforcement Learning
- Learning Equational Theorem Proving
- Considerations on Approaches and Metrics in Automated Theorem Generation/Finding in Geometry
- BAIT: Benchmarking (Embedding) Architectures for Interactive Theorem-Proving
- First Experiments with Neural cvc5
- Learning to solve geometric construction problems from images
- Generating Symbolic Reasoning Problems with Transformer GANs
- Faster Smarter Induction in Isabelle/HOL
- Learned Provability Likelihood for Tactical Search
- Lassie: HOL4 Tactics by Example
- Online Machine Learning Techniques for Coq: A Comparison