Adversarial observations for probabilistic state space models for robust reinforcement learning

M. Santos Pascual, D. Ríos Insua

Decision-making under partial or adversarial observability requires accurate inference of the environment’s latent state and its associated uncertainty. This work analyzes 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 remain consistent with the observation model of the underlying decision-making system. We analyze how such adversarial yet realistic observation shifts propagate through the latent state inference process 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.

Palabras clave: Adversarial Machine Learning, Reinforcement Learning, State Space Models, Kalman Filter

Procesos Estocásticos
2 de septiembre de 2026  17:40
Aula 22

I. Armenteros Rojas, A. I. Gallardo Cordovilla, M. Jiménez Medina, A. Pereira das Neves Yedig

M. González Bernal, R. E. Lillo, P. Ramírez-Cobo, L. Senande

A. Guada Azze, B. D'Auria, G. Ferrari

M. Mota Medina, M. Molina Fernández