Distributed Reinforcement Learning for Decentralized Linear Quadratic Control: A Derivative-Free Policy Optimization Approach.
Yingying LiYujie TangRunyu ZhangNa LiPublished in: L4DC (2020)
Keyphrases
- derivative free
- optimal control
- linear quadratic
- reinforcement learning
- unconstrained optimization
- neural learning
- gradient method
- optimization methods
- closed loop
- control system
- vector valued
- control strategy
- constrained optimization
- fan beam
- learning algorithm
- dynamical systems
- machine learning
- dynamic programming
- optimization algorithm
- step size
- convergence rate
- optimization method
- optimization problems
- multiscale
- neural network