A Reinforcement Learning based Reset Policy for CDCL SAT Solvers.
Chunxiao LiCharlie LiuJonathan ChungZhengyang LuPiyush JhaVijay GaneshPublished in: CoRR (2024)
Keyphrases
- sat solvers
- reinforcement learning
- optimal policy
- orders of magnitude
- constraint satisfaction
- markov decision process
- max sat
- propositional satisfiability
- sat solving
- sat instances
- boolean satisfiability
- search space
- state space
- sat problem
- partially observable
- search strategies
- dynamic programming
- search tree
- markov decision processes
- partially observable markov decision processes
- decision problems
- clause learning
- information retrieval
- optimal solution