Pseudo-Rehearsal: Achieving Deep Reinforcement Learning without Catastrophic Forgetting.
Craig AtkinsonBrendan McCaneLech SzymanskiAnthony V. RobinsPublished in: CoRR (2018)
Keyphrases
- reinforcement learning
- learning algorithm
- model free
- function approximation
- optimal policy
- reinforcement learning algorithms
- robotic control
- state space
- incremental learning
- transfer learning
- policy search
- multi agent reinforcement learning
- temporal difference
- optimal control
- markov decision processes
- multi agent
- dynamic programming
- action selection
- action space
- information retrieval
- neural network
- data sets
- hidden markov models
- learning process
- artificial intelligence
- reinforcement learning methods
- machine learning