Constrained Reinforcement Learning via Dissipative Saddle Flow Dynamics.
Tianqi ZhengPengcheng YouEnrique MalladaPublished in: IEEECONF (2022)
Keyphrases
- reinforcement learning
- function approximation
- dynamical systems
- dynamic model
- state space
- fluid flow
- robotic control
- machine learning
- flow patterns
- temporal difference learning
- partially observable
- optimal control
- learning problems
- temporal difference
- temporal evolution
- action selection
- blood flow
- recurrent networks
- critical points