Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggers

Chace Ashcraft; Kiran Karra

Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggers

Machine Learning 2021-06-16 v1 Cryptography and Security

Authors: Chace Ashcraft , Kiran Karra

Abstract

In this paper, we propose a new data poisoning attack and apply it to deep reinforcement learning agents. Our attack centers on what we call in-distribution triggers, which are triggers native to the data distributions the model will be trained on and deployed in. We outline a simple procedure for embedding these, and other, triggers in deep reinforcement learning agents following a multi-task learning paradigm, and demonstrate in three common reinforcement learning environments. We believe that this work has important implications for the security of deep learning models.

Keywords

data poisoning attack reinforcement learning deep learning

Cite

@article{arxiv.2106.07798,
  title  = {Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggers},
  author = {Chace Ashcraft and Kiran Karra},
  journal= {arXiv preprint arXiv:2106.07798},
  year   = {2021}
}

Comments

4 pages, 1 figure, Published at ICLR 2021 Workshop on Security and Safety in Machine Learning Systems

Related papers

View all related →