Reward Function Using Inverse Reinforcement Learning and Fuzzy Reasoning.
Yuta KatoMasayoshi KanohTsuyoshi NakamuraPublished in: SCIS/ISIS (2020)
Keyphrases
- fuzzy reasoning
- inverse reinforcement learning
- reward function
- fuzzy logic
- markov decision processes
- reinforcement learning
- preference elicitation
- fuzzy sets
- state space
- multiple agents
- fuzzy rules
- membership functions
- reinforcement learning algorithms
- partially observable
- transition probabilities
- state variables
- optimal policy
- generative model
- simple examples
- neural network
- rough sets
- learning agent
- em algorithm
- control policies
- genetic algorithm