RL-CD: Dealing with Non-Stationarity in Reinforcement Learning.
Bruno Castro da SilvaEduardo W. BassoAna L. C. BazzanPaulo Martins EngelPublished in: AAAI (2006)
Keyphrases
- reinforcement learning
- function approximation
- model free
- temporal difference
- state space
- non stationary
- rl algorithms
- reinforcement learning algorithms
- markov decision processes
- action selection
- learning algorithm
- control problems
- machine learning
- optimal control
- transfer learning
- optimal policy
- supervised learning
- dynamic programming
- function approximators
- approximate dynamic programming
- continuous state
- temporal difference learning
- learning problems
- multi agent
- learning process
- action space
- markov decision process
- state action
- learning agents
- partially observable markov decision processes
- neural network
- markov decision problems
- actor critic
- exploration strategy
- partially observable domains
- continuous state and action spaces