SRL: Scaling Distributed Reinforcement Learning to Over Ten Thousand Cores.
Zhiyu MeiWei FuJiaxuan GaoGuangju WangHuanchen ZhangYi WuPublished in: ICLR (2024)
Keyphrases
- reinforcement learning
- multi agent
- distributed systems
- distributed environment
- machine learning
- function approximation
- learning algorithm
- cooperative
- peer to peer
- database systems
- tens of thousands
- learning strategies
- model free
- heterogeneous environments
- temporal difference
- reinforcement learning algorithms
- data sets
- distributed data
- optimal control
- computer networks
- dynamic programming
- databases