IMPALA: Scalable Distributed Deep-RL with Importance Weighted Actor-Learner Architectures.
Lasse EspeholtHubert SoyerRémi MunosKaren SimonyanVolodymyr MnihTom WardYotam DoronVlad FiroiuTim HarleyIain DunningShane LeggKoray KavukcuogluPublished in: CoRR (2018)
Keyphrases
- scalable distributed
- reinforcement learning
- learning process
- learning environment
- multi agent
- relative importance
- file system
- social learning
- e learning
- reinforcement learning algorithms
- model free
- state space
- data sets
- learning materials
- adaptive learning
- markov decision processes
- transfer learning
- scheduling problem
- order statistics