paper-with-me

Papers

Adversarial observations in probabilistic State-Space Models for robust Reinforcement Learning

2026-06-18 · M. Santos-Pascual, D. Ríos Insua arxiv

Decision-making under partial or adversarial observability requires accurate inference of the environment's latent state and its associated uncertainty. This work analyses adversarial attacks on linear probabilistic state-space models, commonly integrated within reinforcement learning architectures, where the attacker alters observations under likelihood constraints that ensure the perturbations remains consistent. We analyze how such adversarial yet realistic observation shifts influence the latent state and influence policy decisions. This perspective provides a principled pathway toward building more robust reinforcement learning systems, with direct relevance to safety-critical domains such as robotics, where reliable operation under sensor noise, partial failures, and adversarial conditions is essential.

📄 PDF Abstract BibTeX arXiv:2606.20880

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

Probabilistic Forecasting with Generative Networks via Scoring Rule Minimization

2021-12-15 · Lorenzo Pacchiardi, Rilwan Adewoyin, Peter Dueben, Ritabrata Dutta

Probabilistic forecasting relies on past observations to provide a probability distribution for a future outcome, which is often evaluated against the realization using a scoring rule. Here, we perform probabilistic fore…

scoring ruleUncertainty QuantificationWeather Forecasting

Understanding Adversarial Attacks on Observations in Deep Reinforcement Learning

2021-06-30 · You Qiaoben, Chengyang Ying, Xinning Zhou, Hang Su 외

Deep reinforcement learning models are vulnerable to adversarial attacks that can decrease a victim's cumulative expected reward by manipulating the victim's observations. Despite the efficiency of previous optimization-…

Deep Reinforcement LearningMuJoCoreinforcement-learningReinforcement Learning+1

Probabilistic Perspectives on Error Minimization in Adversarial Reinforcement Learning

2024-06-07 · Roman Belaire, Arunesh Sinha, Pradeep Varakantham

Deep Reinforcement Learning (DRL) policies are highly susceptible to adversarial noise in observations, which poses significant risks in safety-critical scenarios. For instance, a self-driving car could experience catast…

counterfactualDeep Reinforcement Learningreinforcement-learningReinforcement Learning

Mitigating Adversarial Perturbations for Deep Reinforcement Learning via Vector Quantization

2024-10-04 · Tung M. Luu, Thanh Nguyen, Tee Joshua Tian Jin, Sungwoon Kim 외

Recent studies reveal that well-performing reinforcement learning (RL) agents in training often lack resilience against adversarial perturbations during deployment. This highlights the importance of building a robust age…

Deep Reinforcement LearningQuantizationreinforcement-learningReinforcement Learning+1

Adversarial Policies: Attacking Deep Reinforcement Learning

2019-05-25 · ICLR 2020 1 · Adam Gleave, Michael Dennis, Cody Wild, Neel Kant 외

Deep reinforcement learning (RL) policies are known to be vulnerable to adversarial perturbations to their observations, similar to adversarial examples for classifiers. However, an attacker is not usually able to direct…

Deep Reinforcement Learningreinforcement-learningReinforcement LearningReinforcement Learning (RL)