mlsanityData to 14 Oct 2026

Investigating the Treacherous Turn in Deep Reinforcement Learning

arXiv 2504.08943 ↗First posted 11 Apr 2025cs.LG cs.AIcited by 1

On the map: Deep reinforcement learning methods → Adversarial robustness and RL in finance

Abstract (arXiv)

The Treacherous Turn refers to the scenario where an artificial intelligence (AI) agent subtly, and perhaps covertly, learns to perform a behavior that benefits itself but is deemed undesirable and potentially harmful to a human supervisor. During training, the agent learns to behave as expected by the human supervisor, but when deployed to perform its task, it performs an alternate behavior without the supervisor there to prevent it. Initial experiments applying DRL to an implementation of the A Link to the Past example do not produce the treacherous turn effect naturally, despite various modifications to the environment intended to produce it. However, in this work, we find the treacherous behavior to be reproducible in a DRL agent when using other trojan injection strategies. This approach deviates from the prototypical treacherous turn behavior since the behavior is explicitly trained into the agent, rather than occurring as an emergent consequence of environmental complexity or poor objective specification. Nonetheless, these experiments provide new insights into the challenges of producing agents capable of true treacherous turn behavior.

Key facts. Cited by 1 paper in the map (data through 14 Oct 2026). Placed in Adversarial robustness and RL in finance, one of 15 directions in Deep reinforcement learning methods. Builds on 4 of its 17 references that are in the map. First posted on arXiv 11 Apr 2025.

Builds on

4 of 17 references.

References · 4

  1. 2017Proximal Policy Optimization Algorithmscited by 15,215
  2. 2017BadNets: Identifying Vulnerabilities in the Machine Learning Model Supply Chaincited by 1,106
  3. 2019TrojDRL: Trojan Attacks on Deep Reinforcement Learning Agentscited by 30
  4. 2021Poisoning Deep Reinforcement Learning Agents with In-Distribution Triggerscited by 20

Built on by

Cited by year

Citing papers per year: 2025: 1.

1 paper.

  1. 2025AI Awarenesscited by 7

Similar, not cited

Close in wording, no citation either way

  1. 2024RAT: Adversarial Attacks on Deep Reinforcement Agents for Targeted Behaviors
  2. 2026BehaviorGuard: Online Backdoor Defense for Deep Reinforcement Learning
  3. 2026Beware Untrusted Simulators -- Reward-Free Backdoor Attacks in Reinforcement Learning
  4. 2022Security of Deep Reinforcement Learning for Autonomous Driving: A Survey
  5. 2024Mitigating Deep Reinforcement Learning Backdoors in the Neural Activation Space
  6. 2026SoK: The Pitfalls of Deep Reinforcement Learning for Cybersecurity
  7. 2026Angel or Demon: Investigating the Plasticity Interventions' Impact on Backdoor Threats in Deep Reinforcement Learning
  8. 2025Beyond Training-time Poisoning: Component-level and Post-training Backdoors in Deep Reinforcement Learning