Conformal Uncertainty Indicator for Continual Test-Time Adaptation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Lyu, Fan, Zhao, Hanyu, Shi, Ziqi, Liu, Ye, Hu, Fuyuan, Zhang, Zhang, Wang, Liang
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866917913714229248
author Lyu, Fan
Zhao, Hanyu
Shi, Ziqi
Liu, Ye
Hu, Fuyuan
Zhang, Zhang
Wang, Liang
author_facet Lyu, Fan
Zhao, Hanyu
Shi, Ziqi
Liu, Ye
Hu, Fuyuan
Zhang, Zhang
Wang, Liang
contents Continual Test-Time Adaptation (CTTA) aims to adapt models to sequentially changing domains during testing, relying on pseudo-labels for self-adaptation. However, incorrect pseudo-labels can accumulate, leading to performance degradation. To address this, we propose a Conformal Uncertainty Indicator (CUI) for CTTA, leveraging Conformal Prediction (CP) to generate prediction sets that include the true label with a specified coverage probability. Since domain shifts can lower the coverage than expected, making CP unreliable, we dynamically compensate for the coverage by measuring both domain and data differences. Reliable pseudo-labels from CP are then selectively utilized to enhance adaptation. Experiments confirm that CUI effectively estimates uncertainty and improves adaptation performance across various existing CTTA methods.
format Preprint
id arxiv_https___arxiv_org_abs_2502_02998
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Conformal Uncertainty Indicator for Continual Test-Time Adaptation
Lyu, Fan
Zhao, Hanyu
Shi, Ziqi
Liu, Ye
Hu, Fuyuan
Zhang, Zhang
Wang, Liang
Machine Learning
Continual Test-Time Adaptation (CTTA) aims to adapt models to sequentially changing domains during testing, relying on pseudo-labels for self-adaptation. However, incorrect pseudo-labels can accumulate, leading to performance degradation. To address this, we propose a Conformal Uncertainty Indicator (CUI) for CTTA, leveraging Conformal Prediction (CP) to generate prediction sets that include the true label with a specified coverage probability. Since domain shifts can lower the coverage than expected, making CP unreliable, we dynamically compensate for the coverage by measuring both domain and data differences. Reliable pseudo-labels from CP are then selectively utilized to enhance adaptation. Experiments confirm that CUI effectively estimates uncertainty and improves adaptation performance across various existing CTTA methods.
title Conformal Uncertainty Indicator for Continual Test-Time Adaptation
topic Machine Learning
url https://arxiv.org/abs/2502.02998