arXiv Artificial Intelligence

SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning

7 days ago 03.09.2026 07:00 👁 2 views
SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning

Quick summary

arXiv:2511.09681v3 Announce Type: replace-cross Abstract: Visual reinforcement learning has achieved remarkable progress in visual control and robotics, but its vulnerability to adversarial perturbations remains underexplored. Most existing black-box attacks focus on vector-based or discrete-action RL, and their effectiveness on image-based continuous control is limited by the large action space and excessive environment queries. We propose SEBA, a sample-efficient framework for black-box adversarial attacks on visual RL agents. SEBA integrates a shadow Q model that estimates cumulative reward

Key takeaways

  • arXiv:2511.09681v3 Announce Type: replace-cross Abstract: Visual reinforcement learning has achieved remarkable progress in visual control and robotics, but its vulnerability to adversarial perturbations remains underexplored.
  • Most existing black-box attacks focus on vector-based or discrete-action RL, and their effectiveness on image-based continuous control is limited by the large action space and excessive environment queries.
  • We propose SEBA, a sample-efficient framework for black-box adversarial attacks on visual RL agents.

Why it matters

“SEBA: Sample-Efficient Black-Box Attacks on Visual Reinforcement Learning” shows why AI risk cannot be reduced to answer accuracy. Access controls, logging, human approval and incident response need to be designed into the workflow from the start.

Kaynak sitede devamını oku: arXiv Artificial Intelligence ↗