A category-theoretic analysis represents conformal prediction as morphisms in stability and measurability categories, proves a commuting diagram decomposition into predictive distribution extraction and region derivation, and shows asymptotic compatibility with Bayesian predictive densities plus e-p
Ryan Martin
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
verdicts
UNVERDICTED 4representative citing papers
Choquet risk ranks valid possibilistic inferential models by linking their efficiency to expected performance of induced confidence sets under concentration penalties.
Regularized e-processes add knowledge-based imprecise-probabilistic regularization to e-processes, yielding anytime-valid inference with efficiency gains and possibility-theoretic uncertainty quantification that satisfies the likelihood principle and avoids sure loss.
A review of possibilistic inferential models that deliver strong frequentist reliability and conditional imprecise-probabilistic reasoning, plus a generalization connecting them to bootstrap and conformal prediction methods.
citing papers explorer
-
A Category-Theoretic Analysis of Conformal Prediction
A category-theoretic analysis represents conformal prediction as morphisms in stability and measurability categories, proves a commuting diagram decomposition into predictive distribution extraction and region derivation, and shows asymptotic compatibility with Bayesian predictive densities plus e-p
-
Efficiency of Valid Inferential Models: Choquet-risk Optimal Possibility Measures, and Direct Comparisons
Choquet risk ranks valid possibilistic inferential models by linking their efficiency to expected performance of induced confidence sets under concentration penalties.
-
Regularized e-processes: anytime valid inference with knowledge-based efficiency gains
Regularized e-processes add knowledge-based imprecise-probabilistic regularization to e-processes, yielding anytime-valid inference with efficiency gains and possibility-theoretic uncertainty quantification that satisfies the likelihood principle and avoids sure loss.
-
Possibilistic inferential models: a review
A review of possibilistic inferential models that deliver strong frequentist reliability and conditional imprecise-probabilistic reasoning, plus a generalization connecting them to bootstrap and conformal prediction methods.