WarpDrive: Extremely Fast End-to-End Deep Multi-Agent Reinforcement Learning on a GPU.
Tian LanSunil SrinivasaStephan ZhengPublished in: CoRR (2021)
Keyphrases
- end to end
- multi agent reinforcement learning
- multi agent
- real time
- multi agent systems
- congestion control
- multi agent learning
- stochastic games
- admission control
- reinforcement learning
- cooperative
- learning agents
- scalable video
- expert systems
- real world
- learning agent
- motion estimation
- general purpose
- internet protocol