Beyond Worst-case Attacks: Robust RL with Adaptive Defense via Non-dominated Policies.
Xiangyu LiuChenghao DengYanchao SunYongyuan LiangFurong HuangPublished in: CoRR (2024)
Keyphrases
- worst case
- optimal policy
- reinforcement learning
- image watermarking
- adaptive control
- defense mechanisms
- lower bound
- network security
- watermarking technique
- average case
- computational complexity
- multi agent
- intrusion detection
- function approximation
- countermeasures
- watermarking algorithm
- dynamic programming
- np hard
- special case
- ddos attacks
- digital image watermarking