Evolve To Control: Evolution-based Soft Actor-Critic for Scalable Reinforcement Learning.
Karush SuriXiao Qi ShiKonstantinos N. PlataniotisYuri A. LawryshynPublished in: CoRR (2020)
Keyphrases
- actor critic
- reinforcement learning
- optimal control
- control problems
- temporal difference
- policy gradient
- approximate dynamic programming
- reinforcement learning algorithms
- function approximation
- control strategy
- control strategies
- gradient method
- control method
- adaptive control
- neuro fuzzy
- action selection
- policy iteration
- control policy
- dynamic programming
- control system
- model free
- multi agent
- average reward
- rl algorithms
- state space
- long run
- step size
- dynamical systems
- optimal policy
- linear programming
- lyapunov function
- supervised learning
- learning algorithm