From Channel Bias to Feature Redundancy: Uncovering the "Less is More" Principle in Few-Shot Learning

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Zhang, Ji, Luo, Xu, Gao, Lianli, Zou, Difan, Shen, Hengtao, Song, Jingkuan
Natura: Preprint
Pubblicazione: 2023
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911145997107200
author Zhang, Ji
Luo, Xu
Gao, Lianli
Zou, Difan
Shen, Hengtao
Song, Jingkuan
author_facet Zhang, Ji
Luo, Xu
Gao, Lianli
Zou, Difan
Shen, Hengtao
Song, Jingkuan
contents Deep neural networks often fail to adapt representations to novel tasks under distribution shifts, especially when only a few examples are available. This paper identifies a core obstacle behind this failure: channel bias, where networks develop a rigid emphasis on feature dimensions that were discriminative for the source task, but this emphasis is misaligned and fails to adapt to the distinct needs of a novel task. This bias leads to a striking and detrimental consequence: feature redundancy. We demonstrate that for few-shot tasks, classification accuracy is significantly improved by using as few as 1-5% of the most discriminative feature dimensions, revealing that the vast majority are actively harmful. Our theoretical analysis confirms that this redundancy originates from confounding feature dimensions-those with high intra-class variance but low inter-class separability-which are especially problematic in low-data regimes. This "less is more" phenomenon is a defining characteristic of the few-shot setting, diminishing as more samples become available. To address this, we propose a simple yet effective soft-masking method, Augmented Feature Importance Adjustment (AFIA), which estimates feature importance from augmented data to mitigate the issue. By establishing the cohesive link from channel bias to its consequence of extreme feature redundancy, this work provides a foundational principle for few-shot representation transfer and a practical method for developing more robust few-shot learning algorithms.
format Preprint
id arxiv_https___arxiv_org_abs_2310_03843
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle From Channel Bias to Feature Redundancy: Uncovering the "Less is More" Principle in Few-Shot Learning
Zhang, Ji
Luo, Xu
Gao, Lianli
Zou, Difan
Shen, Hengtao
Song, Jingkuan
Computer Vision and Pattern Recognition
Machine Learning
Deep neural networks often fail to adapt representations to novel tasks under distribution shifts, especially when only a few examples are available. This paper identifies a core obstacle behind this failure: channel bias, where networks develop a rigid emphasis on feature dimensions that were discriminative for the source task, but this emphasis is misaligned and fails to adapt to the distinct needs of a novel task. This bias leads to a striking and detrimental consequence: feature redundancy. We demonstrate that for few-shot tasks, classification accuracy is significantly improved by using as few as 1-5% of the most discriminative feature dimensions, revealing that the vast majority are actively harmful. Our theoretical analysis confirms that this redundancy originates from confounding feature dimensions-those with high intra-class variance but low inter-class separability-which are especially problematic in low-data regimes. This "less is more" phenomenon is a defining characteristic of the few-shot setting, diminishing as more samples become available. To address this, we propose a simple yet effective soft-masking method, Augmented Feature Importance Adjustment (AFIA), which estimates feature importance from augmented data to mitigate the issue. By establishing the cohesive link from channel bias to its consequence of extreme feature redundancy, this work provides a foundational principle for few-shot representation transfer and a practical method for developing more robust few-shot learning algorithms.
title From Channel Bias to Feature Redundancy: Uncovering the "Less is More" Principle in Few-Shot Learning
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2310.03843