Improving Coverage in Combined Prediction Sets with Weighted p-values

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wong, Gina, Prinster, Drew, Saria, Suchi, Chellappa, Rama, Liu, Anqi
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914217790013440
author Wong, Gina
Prinster, Drew
Saria, Suchi
Chellappa, Rama
Liu, Anqi
author_facet Wong, Gina
Prinster, Drew
Saria, Suchi
Chellappa, Rama
Liu, Anqi
contents Conformal prediction quantifies the uncertainty of machine learning models by augmenting point predictions with valid prediction sets. For complex scenarios involving multiple trials, models, or data sources, conformal prediction sets can be aggregated to create a prediction set that captures the overall uncertainty, often improving precision. However, aggregating multiple prediction sets with individual $1-α$ coverage inevitably weakens the overall guarantee, typically resulting in $1-2α$ worst-case coverage. In this work, we propose a framework for the weighted aggregation of prediction sets, where weights are assigned to each prediction set based on their contribution. Our framework offers flexible control over how the sets are aggregated, achieving tighter coverage bounds that interpolate between the $1-2α$ guarantee of the combined models and the $1-α$ guarantee of an individual model depending on the distribution of weights. Importantly, our framework generalizes to data-dependent weights, as we derive a procedure for weighted aggregation that maintains finite-sample validity even when the weights depend on the data. This extension makes our framework broadly applicable to settings where weights are learned, such as mixture-of-experts (MoE), and we demonstrate through experiments in the MoE setting that our methods achieve adaptive coverage.
format Preprint
id arxiv_https___arxiv_org_abs_2505_11785
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Improving Coverage in Combined Prediction Sets with Weighted p-values
Wong, Gina
Prinster, Drew
Saria, Suchi
Chellappa, Rama
Liu, Anqi
Machine Learning
Artificial Intelligence
Conformal prediction quantifies the uncertainty of machine learning models by augmenting point predictions with valid prediction sets. For complex scenarios involving multiple trials, models, or data sources, conformal prediction sets can be aggregated to create a prediction set that captures the overall uncertainty, often improving precision. However, aggregating multiple prediction sets with individual $1-α$ coverage inevitably weakens the overall guarantee, typically resulting in $1-2α$ worst-case coverage. In this work, we propose a framework for the weighted aggregation of prediction sets, where weights are assigned to each prediction set based on their contribution. Our framework offers flexible control over how the sets are aggregated, achieving tighter coverage bounds that interpolate between the $1-2α$ guarantee of the combined models and the $1-α$ guarantee of an individual model depending on the distribution of weights. Importantly, our framework generalizes to data-dependent weights, as we derive a procedure for weighted aggregation that maintains finite-sample validity even when the weights depend on the data. This extension makes our framework broadly applicable to settings where weights are learned, such as mixture-of-experts (MoE), and we demonstrate through experiments in the MoE setting that our methods achieve adaptive coverage.
title Improving Coverage in Combined Prediction Sets with Weighted p-values
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2505.11785