Damped Anderson Mixing for Deep Reinforcement Learning: Acceleration, Convergence, and Stabilization.
Ke SunYafei WangYi LiuYingnan ZhaoBo PanShangling JuiBei JiangLinglong KongPublished in: NeurIPS (2021)
Keyphrases
- reinforcement learning
- stochastic approximation
- function approximation
- convergence rate
- state space
- machine learning
- learning algorithm
- global convergence
- optimal policy
- model free
- reinforcement learning algorithms
- markov decision processes
- convergence speed
- transfer learning
- temporal difference
- monte carlo
- dynamic programming
- temporal difference learning
- convergence analysis
- multi agent reinforcement learning
- robotic control
- partially observable
- iterative algorithms
- policy iteration
- deep learning
- multi agent
- newton method