Enumeration and Deduction Driven Co-Synthesis of CCSL Specifications using Reinforcement Learning.
Ming HuJiepin DingMin ZhangFrédéric MalletMingsong ChenPublished in: RTSS (2021)
Keyphrases
- natural language
- reinforcement learning
- written in natural language
- machine learning
- function approximation
- dynamic programming
- knowledge representation
- inference rules
- search space
- data driven
- model free
- texture synthesis
- markov decision processes
- program synthesis
- state space
- stochastic approximation
- reinforcement learning algorithms
- optimal policy
- learning algorithm
- high level
- search algorithm
- robotic control
- formal specification
- theorem proving
- multi agent
- control flow
- temporal difference learning
- software development
- computer algebra
- transition model
- intuitionistic logic
- genetic algorithm