Directed Exploration Via Learnable Probability Distribution For Random Action Selection.
Petros GiannakopoulosAggelos PikrakisYannis CotronisPublished in: ICME (2020)
Keyphrases
- action selection
- probability distribution
- robot soccer
- central limit theorem
- basal ganglia
- reinforcement learning
- decision making
- temporal difference
- random variables
- human robot
- conditional independence
- learning algorithm
- action space
- bayesian networks
- stochastic processes
- reinforcement learning algorithms
- early stage
- information processing
- classification noise
- neural network
- continuous state and action spaces