Causal Based Action Selection Policy for Reinforcement Learning.
Ivan Feliciano-AvelinoArquímides Méndez-MolinaEduardo F. MoralesLuis Enrique SucarPublished in: MICAI (1) (2021)
Keyphrases
- action selection
- reinforcement learning
- robot soccer
- temporal difference
- basal ganglia
- human robot
- function approximation
- decision making
- action space
- continuous state and action spaces
- optimal policy
- causal models
- causal relationships
- learning algorithm
- total reward
- policy evaluation
- reinforcement learning algorithms
- model free
- state space
- bayesian networks
- machine learning
- policy iteration
- function approximators
- control policy
- state action
- natural language
- action selection mechanism