Robustness Verification of Deep Reinforcement Learning Based Control Systems using Reward Martingales.
Dapeng ZhiPeixin WangCheng ChenMin ZhangPublished in: CoRR (2023)
Keyphrases
- reinforcement learning
- control system
- function approximation
- state space
- reinforcement learning algorithms
- eligibility traces
- model free
- control strategies
- markov decision processes
- machine learning
- partially observable environments
- control policies
- reward function
- learning algorithm
- supervised learning
- total reward
- average reward
- real time
- policy gradient
- multi agent
- reinforcement learning methods
- face verification
- optimal policy
- model checking
- dynamic programming
- learning agent
- mobile robot
- long run
- multi armed bandit
- optimal control
- fault detection and isolation
- reward shaping
- transfer learning