Dynamic Action Repetition for Deep Reinforcement Learning.
Aravind S. LakshminarayananSahil SharmaBalaraman RavindranPublished in: AAAI (2017)
Keyphrases
- reinforcement learning
- action space
- action selection
- learning algorithm
- reward shaping
- function approximation
- optimal policy
- partially observable domains
- transition model
- markov decision processes
- real time
- multi agent
- transfer learning
- dynamic environments
- dynamic programming
- video sequences
- continuous state
- case study
- search engine
- data sets