TY  - RPRT
TI  - Provably Invincible Adversarial Attacks on Reinforcement Learning Systems: A Rate-Distortion Information-Theoretic Approach
AU  - Ziqing Lu
AU  - Lifeng Lai
AU  - Weiyu Xu
PY  - 2025
UR  - https://arxiv.org/abs/2510.13792
ID  - 2510.13792
ER  -