Feedback Manipulation Regularization: Enabling Offline Agent Alignment for Imitation Learning

Jul 10, 2026

Researchers have introduced Feedback Manipulation Regularization (FMR), an algorithm-agnostic method that leverages evaluative feedback to improve the alignment of imitation learning policies in sequential decision-making tasks. In experiments using Safety Gymnasium environments, FMR demonstrated up to a 98% reduction in misalignment across various imitation learning algorithms and maintained robustness even with limited data.

Why it matters: FMR provides a single-stage offline training approach that effectively integrates demonstrations and feedback for agent alignment, addressing limitations of existing multi-stage methods.

Full story at: arXiv AI/ML ↗