Stacked conformal prediction
This work addresses the challenge of efficient and valid uncertainty quantification for ensemble methods in machine learning, but it appears incremental as it builds on existing conformal prediction techniques.
The paper tackled the problem of conformalizing stacked ensembles of predictive models, achieving approximate marginal validity without needing a separate calibration sample and showing favorable empirical comparison to a standard inductive alternative.
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.