Probabilistic policy reuse in a reinforcement learning agent.
Fernando FernándezManuela M. VelosoPublished in: AAMAS (2006)
Keyphrases
- learning agent
- reinforcement learning
- selective perception
- optimal policy
- reward function
- bayesian networks
- state space
- agent learns
- learning algorithm
- solving problems
- probabilistic model
- reinforcement learning algorithms
- learning capabilities
- function approximation
- generative model
- action selection
- markov decision processes
- learning process
- machine learning
- transfer learning
- uncertain data
- online learning
- temporal difference
- search space