CopyCAT: Taking Control of Neural Policies with Constant Attacks
We propose a new perspective on adversarial attacks against deep reinforcement learning agents. Our main contribution is CopyCAT, a targeted attack able to consistently lure an agent into following an outsider's policy. It is pre-computed, therefore fast inferred, and could thus be usable in a real-time scenario. We show its effectiveness on Atari 2600 games in the novel read-only setting. In this setting, the adversary cannot directly modify the agent's state -- its representation of the environment -- but can only attack the agent's observation -- its perception of the environment. Directly modifying the agent's state would require a write-access to the agent's inner workings and we argue that this assumption is too strong in realistic settings.
Code (0)
등록된 구현이 없습니다.
Tasks
Atari GamesDeep Reinforcement Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)Similar Papers 제목 키워드 기반
Copycat CNN: Stealing Knowledge by Persuading Confession with Random Non-Labeled Data
In the past few years, Convolutional Neural Networks (CNNs) have been achieving state-of-the-art performance on a variety of problems. Many companies employ resources and money to generate these models and provide them a…
COPYCAT: Practical Adversarial Attacks on Visualization-Based Malware Detection
Despite many attempts, the state-of-the-art of adversarial machine learning on malware detection systems generally yield unexecutable samples. In this work, we set out to examine the robustness of visualization-based mal…
Adversarial AttackMalware DetectionThe Impact of Load Altering Attacks on Distribution Systems with ZIP Loads
Load-altering attacks (LAAs) pose a significant threat to power systems with Internet of Things (IoT)-controllable load devices. This research examines the detrimental impact of LAAs on the voltage profile of distributio…
Fighting Copycat Agents in Behavioral Cloning from Observation Histories
Imitation learning trains policies to map from input observations to the actions that an expert would choose. In this setting, distribution shift frequently exacerbates the effect of misattributing expert actions to nuis…
Imitation LearningReimplementation and Reinterpretation of the Copycat Project
We present the reinterpreted and reimplemented Copycat project, an architecture solving letter analogy domain problems. To support a flexible implementation change and rigor testing process, we propose a implementation m…