Cost-Aware Adaptive Conformal Inference for Runtime Assurance in Dynamic Environments
For systems operating in dynamic environments, this work addresses the need for runtime assurance that accounts for varying severity of constraint violations, offering a more nuanced approach than existing methods.
This paper introduces Cost-Aware Adaptive Conformal Inference (ACI), a framework that adapts prediction sets based on violation costs, providing dual guarantees on long-run average violation frequencies and cumulative violation cost. Experiments validate its effectiveness for online risk-aware controller synthesis.
This paper addresses the problem of providing runtime assurance for systems operating online under unknown and potentially time-varying data distributions. We propose Cost-Aware Adaptive Conformal Inference (ACI), a novel framework that incorporates constraint violation costs directly into the conformal adaptation mechanism. Our key insight is that uncertainty margins should adapt not only to the frequency of constraint violations but also to their severity. We formalize this through a cost-aware loss function that couples the miscoverage indicator with violation costs. Unlike existing methods that regulate a single controlled metric, our approach provides a dual statistical guarantee: simultaneously bounding the long-run average violation frequencies (reliability) and cumulative violation cost (harm). By weighting prediction failures according to their severity, the algorithm enables the controller to respond proportionally to violation severity, expanding prediction sets aggressively when necessary while maintaining efficiency during nominal operation. We integrate Cost-Aware ACI into a robust control synthesis framework, creating a closed-loop system that balances task performance with runtime risk control without requiring explicit model knowledge. Experiments validate its effectiveness for online risk-aware controller synthesis.