Compositional Reinforcement Learning from Logical Specifications.
Kishor JothimuruganSuguman BansalOsbert BastaniRajeev AlurPublished in: CoRR (2021)
Keyphrases
- reinforcement learning
- function approximation
- optimal policy
- state space
- temporal difference
- reinforcement learning algorithms
- high level
- temporal difference learning
- reinforcement learning methods
- learning algorithm
- machine learning
- markov decision processes
- specification language
- robotic control
- multi agent reinforcement learning
- autonomous learning
- policy search
- logical representation
- delay insensitive
- functional requirements
- optimal control
- formal specification
- database
- learning problems
- supervised learning
- e learning
- neural network
- model free
- logical framework
- multi agent
- case study
- reactive systems
- web services
- information systems
- logical reasoning
- logical rules
- databases