Online Learning of Deceptive Policies under Intermittent Observation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909797087969280 |
|---|---|
| author | Puthumanaillam, Gokul Padmanabhan, Ram Fuentes, Jose Cruz, Nicole Padrao, Paulo Hernandez, Ruben Jiang, Hao Schafer, William Bobadilla, Leonardo Ornik, Melkior |
| author_facet | Puthumanaillam, Gokul Padmanabhan, Ram Fuentes, Jose Cruz, Nicole Padrao, Paulo Hernandez, Ruben Jiang, Hao Schafer, William Bobadilla, Leonardo Ornik, Melkior |
| contents | In supervisory control settings, autonomous systems are not monitored continuously. Instead, monitoring often occurs at sporadic intervals within known bounds. We study the problem of deception, where an agent pursues a private objective while remaining plausibly compliant with a supervisor's reference policy when observations occur. Motivated by the behavior of real, human supervisors, we situate the problem within Theory of Mind: the representation of what an observer believes and expects to see. We show that Theory of Mind can be repurposed to steer online reinforcement learning (RL) toward such deceptive behavior. We model the supervisor's expectations and distill from them a single, calibrated scalar -- the expected evidence of deviation if an observation were to happen now. This scalar combines how unlike the reference and current action distributions appear, with the agent's belief that an observation is imminent. Injected as a state-dependent weight into a KL-regularized policy improvement step within an online RL loop, this scalar informs a closed-form update that smoothly trades off self-interest and compliance, thus sidestepping hand-crafted or heuristic policies. In real-world, real-time hardware experiments on marine (ASV) and aerial (UAV) navigation, our ToM-guided RL runs online, achieves high return and success with observed-trace evidence calibrated to the supervisor's expectations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_14453 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Online Learning of Deceptive Policies under Intermittent Observation Puthumanaillam, Gokul Padmanabhan, Ram Fuentes, Jose Cruz, Nicole Padrao, Paulo Hernandez, Ruben Jiang, Hao Schafer, William Bobadilla, Leonardo Ornik, Melkior Robotics Multiagent Systems Systems and Control In supervisory control settings, autonomous systems are not monitored continuously. Instead, monitoring often occurs at sporadic intervals within known bounds. We study the problem of deception, where an agent pursues a private objective while remaining plausibly compliant with a supervisor's reference policy when observations occur. Motivated by the behavior of real, human supervisors, we situate the problem within Theory of Mind: the representation of what an observer believes and expects to see. We show that Theory of Mind can be repurposed to steer online reinforcement learning (RL) toward such deceptive behavior. We model the supervisor's expectations and distill from them a single, calibrated scalar -- the expected evidence of deviation if an observation were to happen now. This scalar combines how unlike the reference and current action distributions appear, with the agent's belief that an observation is imminent. Injected as a state-dependent weight into a KL-regularized policy improvement step within an online RL loop, this scalar informs a closed-form update that smoothly trades off self-interest and compliance, thus sidestepping hand-crafted or heuristic policies. In real-world, real-time hardware experiments on marine (ASV) and aerial (UAV) navigation, our ToM-guided RL runs online, achieves high return and success with observed-trace evidence calibrated to the supervisor's expectations. |
| title | Online Learning of Deceptive Policies under Intermittent Observation |
| topic | Robotics Multiagent Systems Systems and Control |
| url | https://arxiv.org/abs/2509.14453 |