Modular and Adaptive Conformal Prediction for Sequential Models via Residual Decomposition
This work addresses the need for robust and interpretable uncertainty quantification in sequential modeling pipelines, particularly for practitioners dealing with distribution shifts in domains like supply chains and finance.
The paper tackles the problem of conformal prediction for sequential models by introducing a framework that decomposes prediction residuals into stage-specific components, enabling uncertainty attribution to specific pipeline stages. Experiments on synthetic and real-world data show the approach maintains coverage under conditions that degrade standard methods while providing interpretable uncertainty attribution.
Conformal prediction offers finite-sample coverage guarantees under minimal assumptions. However, existing methods treat the entire modeling process as a black box, overlooking opportunities to exploit modular structure. We introduce a conformal prediction framework for two-stage sequential models, where an upstream predictor generates intermediate representations for a downstream model. By decomposing the overall prediction residual into stage-specific components, our method enables practitioners to attribute uncertainty to specific pipeline stages. We develop a risk-controlled parameter selection procedure using family-wise error rate (FWER) control to calibrate stage-wise scaling parameters, and propose an adaptive extension for non-stationary settings that preserves long-run coverage guarantees. Experiments on synthetic distribution shifts, as well as real-world supply chain and stock market data, demonstrate that our approach maintains coverage under conditions that degrade standard conformal methods, while providing interpretable stage-wise uncertainty attribution. This framework offers diagnostic advantages and robust coverage that standard conformal methods lack.