Distributed lifelong reinforcement learning with sub-linear regret.
Rasul TutunovJulia El ZiniHaitham Bou-AmmarAli JadbabaiePublished in: CDC (2017)
Keyphrases
- reinforcement learning
- multi agent
- distributed environment
- distributed systems
- function approximation
- lower bound
- machine learning
- cooperative
- total reward
- learning process
- state space
- function approximators
- confidence bounds
- expert advice
- model free
- computing environments
- learning experience
- mobile agents
- dynamic programming
- markov decision processes
- computer networks
- loss function
- online learning
- action selection
- reinforcement learning algorithms
- peer to peer
- multi agent systems
- objective function