Hysteresis-Based RL: Robustifying Reinforcement Learning-based Control Policies via Hybrid Control.
Jan de PriesterRicardo G. SanfeliceNathan van de WouwPublished in: CoRR (2022)
Keyphrases
- control policies
- reinforcement learning
- control policy
- optimal policy
- action space
- continuous state
- function approximation
- control strategies
- state space
- reinforcement learning algorithms
- motion control
- markov decision processes
- multi agent
- control system
- action selection
- dynamic programming
- long run
- model free
- temporal difference
- machine learning
- finite horizon
- decision problems
- markov decision process
- real time
- learning algorithm
- dynamic environments
- reward function
- multistage
- genetic algorithm