Test-Time Adaptation with Principal Component Analysis
Abstract
Machine Learning models are prone to fail when test data are different from training data, a situation often encountered in real applications known as distribution shift. While still valid, the training-time knowledge becomes less effective, requiring a test-time adaptation to maintain high performance. Following approaches that assume batch-norm layer and use their statistics for adaptation, we propose a Test-Time Adaptation with Principal Component Analysis (TTAwPCA), which presumes a fitted PCA and adapts at test time a spectral filter based on the singular values of the PCA for robustness to corruptions. TTAwPCA combines three components: the output of a given layer is decomposed using a Principal Component Analysis (PCA), filtered by a penalization of its singular values, and reconstructed with the PCA inverse transform. This generic enhancement adds fewer parameters than current methods. Experiments on CIFAR-10-C and CIFAR- 100-C demonstrate the effectiveness and limits of our method using a unique filter of 2000 parameters.
Cite
@article{arxiv.2209.05779,
title = {Test-Time Adaptation with Principal Component Analysis},
author = {Thomas Cordier and Victor Bouvier and Gilles Hénaff and Céline Hudelot},
journal= {arXiv preprint arXiv:2209.05779},
year = {2022}
}
Comments
7 pages, 2 figures, 2 tables, accepted at Workshop on Trustworthy Artificial Intelligence in conjunction with ECML/PKDD 22