BanditFuzz: A Reinforcement-Learning Based Performance Fuzzer for SMT Solvers.
Joseph ScottFederico MoraVijay GaneshPublished in: VSTTE (2020)
Keyphrases
- reinforcement learning
- modulo theories
- function approximation
- state space
- satisfiability modulo theories
- multi agent
- markov decision processes
- robotic control
- learning algorithm
- policy search
- statistical machine translation
- reinforcement learning algorithms
- answer set programming
- optimal policy
- machine learning
- temporal difference
- model free
- learning problems
- action selection
- sat solving