SOURCE-LINKED INTELLIGENCE

Robust Policy Optimization via Adversarial Importance Sampling

arXiv · AI, language, vision and robotics · article · Sep 11, 2026 · UTC

Significant progress has been made in safeguarding deep reinforcement learning (DRL) policies against input perturbations. Developing robust DRL involves three main stages: algorithm design, implementation, and evaluation. In this work, we identify and address a key limitation at each stage. First, we introduce Adversarial Importance Sampling (Advis), a method that uses importance sampling over trajectories from standard training to estimate and optimize verifiable worst-case returns. Advis satisfies three desirable criteria not jointly achieved by prior work: it requires no additional environ

Read original source ↗ Open in workspace

recordType

paper

region

Global