Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggers

2021/06/14 by Chace Ashcraft, Ashcraft, Chace, Kiran Karra +1 · 4 citations
Computer Science · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #Cryptography and Security (cs.CR) #FOS: Computer and information sciences #Machine Learning (cs.LG) #Reinforcement Learning in Robotics

paper · pdf · doi:10.48550/arxiv.2106.07798

openalex publication_date 2021/06/14 · openalex created_date 2021/06/22 · openalex updated_date 2026/07/28

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.

Citations

• BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chain
• Targeted Backdoor Attacks on Deep Learning Systems Using Data Poisoning
• Spectral Signatures in Backdoor Attacks
• NeuronInspect: Detecting Backdoors in Neural Networks via Output Explanations
• The TrojAI Software Framework: An OpenSource tool for Embedding Trojans\n into Deep Learning Models
• Weight Poisoning Attacks on Pre-trained Models
• Blind Backdoors in Deep Learning Models
• Natural Backdoor Attack on Text Data
• Invisible Backdoor Attacks on Deep Neural Networks via Steganography and Regularization
• PettingZoo: Gym for Multi-Agent Reinforcement Learning

Cited by

• Exposing Vulnerabilities in RL: A Novel Stealthy Backdoor Attack through Reward Poisoning
• Beyond Training-time Poisoning: Component-level and Post-training Backdoors in Deep Reinforcement Learning
• TrojanTO: Action-Level Backdoor Attacks against Trajectory Optimization Models
• TooBadRL: Trigger Optimization to Boost Effectiveness of Backdoor Attacks on Deep Reinforcement Learning

Related

• Backdoors in DRL: Four Environments Focusing on In-distribution Triggers
• Efficient Reward Poisoning Attacks on Online Deep Reinforcement Learning
• TrojDRL: Trojan Attacks on Deep Reinforcement Learning Agents
• Execute Order 66: Targeted Data Poisoning for Reinforcement Learning
• Black-Box Targeted Reward Poisoning Attack Against Online Deep Reinforcement Learning