Gossip-based Actor-Learner Architectures for Deep Reinforcement Learning.
Mahmoud AssranJoshua RomoffNicolas BallasJoelle PineauMike RabbatPublished in: CoRR (2019)
Keyphrases
- reinforcement learning
- learning process
- active exploration
- learning environment
- function approximation
- e learning
- learning materials
- language learning
- state space
- reinforcement learning algorithms
- action selection
- adaptive learning
- learner model
- temporal difference learning
- dynamic programming
- machine learning
- iterative learning
- reinforcement learning methods
- robotic control
- autonomous learning
- agent learns
- control policy
- real time
- deep learning
- temporal difference
- learning resources
- learning experience
- optimal policy
- collaborative learning
- multi agent
- decision trees
- metadata
- data sets