Conformal Prediction for Generative Models via Adaptive Cluster-Based Density Estimation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Yang, Qidong, Zhu, Qianyu Julie, Giezendanner, Jonathan, Marzouk, Youssef, Bates, Stephen, Wang, Sherrie
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917233256562688
author Yang, Qidong
Zhu, Qianyu Julie
Giezendanner, Jonathan
Marzouk, Youssef
Bates, Stephen
Wang, Sherrie
author_facet Yang, Qidong
Zhu, Qianyu Julie
Giezendanner, Jonathan
Marzouk, Youssef
Bates, Stephen
Wang, Sherrie
contents Conditional generative models map input variables to complex, high-dimensional distributions, enabling realistic sample generation in a diverse set of domains. A critical challenge with these models is the absence of calibrated uncertainty, which undermines trust in individual outputs for high-stakes applications. To address this issue, we propose a systematic conformal prediction approach tailored to conditional generative models, leveraging density estimation on model-generated samples. We introduce a novel method called CP4Gen, which utilizes clustering-based density estimation to construct prediction sets that are less sensitive to outliers, more interpretable, and of lower structural complexity than existing methods. Extensive experiments on synthetic datasets and real-world applications, including climate emulation tasks, demonstrate that CP4Gen consistently achieves superior performance in terms of prediction set volume and structural simplicity. Our approach offers practitioners a powerful tool for uncertainty estimation associated with conditional generative models, particularly in scenarios demanding rigorous and interpretable prediction sets.
format Preprint
id arxiv_https___arxiv_org_abs_2601_22298
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Conformal Prediction for Generative Models via Adaptive Cluster-Based Density Estimation
Yang, Qidong
Zhu, Qianyu Julie
Giezendanner, Jonathan
Marzouk, Youssef
Bates, Stephen
Wang, Sherrie
Machine Learning
Artificial Intelligence
Atmospheric and Oceanic Physics
Conditional generative models map input variables to complex, high-dimensional distributions, enabling realistic sample generation in a diverse set of domains. A critical challenge with these models is the absence of calibrated uncertainty, which undermines trust in individual outputs for high-stakes applications. To address this issue, we propose a systematic conformal prediction approach tailored to conditional generative models, leveraging density estimation on model-generated samples. We introduce a novel method called CP4Gen, which utilizes clustering-based density estimation to construct prediction sets that are less sensitive to outliers, more interpretable, and of lower structural complexity than existing methods. Extensive experiments on synthetic datasets and real-world applications, including climate emulation tasks, demonstrate that CP4Gen consistently achieves superior performance in terms of prediction set volume and structural simplicity. Our approach offers practitioners a powerful tool for uncertainty estimation associated with conditional generative models, particularly in scenarios demanding rigorous and interpretable prediction sets.
title Conformal Prediction for Generative Models via Adaptive Cluster-Based Density Estimation
topic Machine Learning
Artificial Intelligence
Atmospheric and Oceanic Physics
url https://arxiv.org/abs/2601.22298