Year: 2025
Type: preprint
Abstract: We consider a method for conformalizing a stacked ensemble of predictive models, showing that the potentially simple form of the meta-learner at the top of the stack enables a procedure with manageable computational cost that achieves approximate marginal validity without requiring the use of a separate calibration sample. Empirical results indicate that the method compares favorably to a standard inductive alternative.
Source: arXiv (Cornell University)
Author: F, Paulo C. Marques
Language: English
Subfield: Statistics and Probability
Field: Mathematics
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W4417301623
Cites:
Cited by:
Related to: