Work

Rolling Conformal Prediction in Sequential Model Training

Year: 2026

Type: preprint

Abstract: We introduce Rolling Conformal Prediction (rolling-CP), a distribution-free predictive inference method for the setting of sequential model training. Specifically, given a data stream $(X_1,Y_1),(X_2,Y_2),\dots$, at each time $n$ the trained model may depend on the observed history $\{(X_i,Y_i)\}_{i

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2609.26951

Authors: Cheng, Chen, Liang, Ruiting, Barber, Rina Foygel

Institutions: University of Illinois Urbana-Champaign, Harvard University, University of Chicago

Language: English

Topic: Data Stream Mining Techniques

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: National Science Foundation, Office of Naval Research

Awards: G2927179718, NSF-Simons AI Institute for the Sky (SkAI Institute), TRIPODS: Institute for Foundations of Data Science

OpenAlex ID: W7214156309

Cites: 12

Cited by:

Related to: 10