TEA: Test-time Energy Adaptation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yuan, Yige, Xu, Bingbing, Hou, Liang, Sun, Fei, Shen, Huawei, Cheng, Xueqi
Format: Preprint
Published: 2023
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911784538996736
author Yuan, Yige
Xu, Bingbing
Hou, Liang
Sun, Fei
Shen, Huawei
Cheng, Xueqi
author_facet Yuan, Yige
Xu, Bingbing
Hou, Liang
Sun, Fei
Shen, Huawei
Cheng, Xueqi
contents Test-time adaptation (TTA) aims to improve model generalizability when test data diverges from training distribution, offering the distinct advantage of not requiring access to training data and processes, especially valuable in the context of large pre-trained models. However, current TTA methods fail to address the fundamental issue: covariate shift, i.e., the decreased generalizability can be attributed to the model's reliance on the marginal distribution of the training data, which may impair model calibration and introduce confirmation bias. To address this, we propose a novel energy-based perspective, enhancing the model's perception of target data distributions without requiring access to training data or processes. Building on this perspective, we introduce $\textbf{T}$est-time $\textbf{E}$nergy $\textbf{A}$daptation ($\textbf{TEA}$), which transforms the trained classifier into an energy-based model and aligns the model's distribution with the test data's, enhancing its ability to perceive test distributions and thus improving overall generalizability. Extensive experiments across multiple tasks, benchmarks and architectures demonstrate TEA's superior generalization performance against state-of-the-art methods. Further in-depth analyses reveal that TEA can equip the model with a comprehensive perception of test distribution, ultimately paving the way toward improved generalization and calibration.
format Preprint
id arxiv_https___arxiv_org_abs_2311_14402
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle TEA: Test-time Energy Adaptation
Yuan, Yige
Xu, Bingbing
Hou, Liang
Sun, Fei
Shen, Huawei
Cheng, Xueqi
Machine Learning
Test-time adaptation (TTA) aims to improve model generalizability when test data diverges from training distribution, offering the distinct advantage of not requiring access to training data and processes, especially valuable in the context of large pre-trained models. However, current TTA methods fail to address the fundamental issue: covariate shift, i.e., the decreased generalizability can be attributed to the model's reliance on the marginal distribution of the training data, which may impair model calibration and introduce confirmation bias. To address this, we propose a novel energy-based perspective, enhancing the model's perception of target data distributions without requiring access to training data or processes. Building on this perspective, we introduce $\textbf{T}$est-time $\textbf{E}$nergy $\textbf{A}$daptation ($\textbf{TEA}$), which transforms the trained classifier into an energy-based model and aligns the model's distribution with the test data's, enhancing its ability to perceive test distributions and thus improving overall generalizability. Extensive experiments across multiple tasks, benchmarks and architectures demonstrate TEA's superior generalization performance against state-of-the-art methods. Further in-depth analyses reveal that TEA can equip the model with a comprehensive perception of test distribution, ultimately paving the way toward improved generalization and calibration.
title TEA: Test-time Energy Adaptation
topic Machine Learning
url https://arxiv.org/abs/2311.14402