machine learning

Distributionally Robust and Safe Imitation Learning

arXiv:2607.13436

summary

The paper introduces a framework that combines Taylor Series Imitation Learning with distributionally robust adaptive control to handle both policy- and uncertainty-induced distribution shifts while enforcing safety constraints, demonstrated on a UAV task.

Abstract

Imitation learning (IL) has achieved remarkable success in complex decision-making tasks. However, its performance is highly sensitive to distribution shifts, which can pose significant safety risks. We propose a distributionally robust and safe IL framework that explicitly addresses both policy-induced and uncertainty-induced distribution shifts. Our approach develops a unified framework leveraging Taylor Series Imitation Learning (TaSIL) to mitigate policy-induced shifts and distributionally robust adaptive control to handle uncertainty-induced shifts. This architecture enables the formulation of an IL problem that optimizes performance under distributional uncertainty while systematically accounting for safety constraints. We demonstrate the effectiveness of the proposed approach on an unmanned aerial vehicle (UAV) case study where the UAV performs a task in an uncertain environment while avoiding unsafe regions.

Topics & keywords

#imitation learning#distributional robustness#safe control#adaptive control#UAVTaylor Series Imitation Learningdistributionally robust optimizationsafety constraintspolicy-induced shiftuncertainty-induced shift
Distributionally Robust and Safe Imitation Learning · wovepaper