AdaptNC: Adaptive Nonconformity Scores for Uncertainty-Aware Autonomous Systems in Dynamic Environments
This addresses the need for more efficient uncertainty quantification in robotics, though it is an incremental improvement over existing online conformal prediction methods.
The paper tackled the problem of overly conservative prediction regions in conformal prediction for autonomous systems under distribution shifts, and proposed AdaptNC to jointly adapt nonconformity scores and thresholds, resulting in significantly reduced prediction region volume while maintaining target coverage.
Rigorous uncertainty quantification is essential for the safe deployment of autonomous systems in unconstrained environments. Conformal Prediction (CP) provides a distribution-free framework for this task, yet its standard formulations rely on exchangeability assumptions that are violated by the distribution shifts inherent in real-world robotics. Existing online CP methods maintain target coverage by adaptively scaling the conformal threshold, but typically employ a static nonconformity score function. We show that this fixed geometry leads to highly conservative, volume-inefficient prediction regions when environments undergo structural shifts. To address this, we propose \textbf{AdaptNC}, a framework for the joint online adaptation of both the nonconformity score parameters and the conformal threshold. AdaptNC leverages an adaptive reweighting scheme to optimize score functions, and introduces a replay buffer mechanism to mitigate the coverage instability that occurs during score transitions. We evaluate AdaptNC on diverse robotic benchmarks involving multi-agent policy changes, environmental changes and sensor degradation. Our results demonstrate that AdaptNC significantly reduces prediction region volume compared to state-of-the-art threshold-only baselines while maintaining target coverage levels.