Scholarlink

<p>This paper presents a novel perspective on enhancing the performance of classification-based Deep Metric Learning (DML). While classification-based DML has seen substantial progress through various regularization techniques, conventional normalization and scaling methods often lead to premature loss saturation and vanishing gradients caused by model overconfidence. To address this issue, we introduce Confidence Control (CC), a new regularization method that prevents this saturation by actively managing prediction confidence. By ensuring sufficient gradient magnitudes throughout training, CC encourages samples to align more strongly with their corresponding class weights. This results in improved feature invariance and tighter intra-class clustering in the embedding space. We propose two implementations of CC: (1) NaiveCC, a direct method that continuously maintains meaningful loss magnitudes by explicitly reducing confidence via a detached logit term; and (2) Evidence-based CC (EVDCC), which addresses the risk of excessive variance reduction (<i>i</i>.<i>e</i>., feature collapse) by imposing a geometric constraint based on eigenvector-based augmentation. Experimental results demonstrate that CC is easily integrable into existing DML frameworks and consistently improves performance, achieving a 2–5% gain in Recall@K.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Confidence Controls Deep Metric Learning

  • 期刊论文

  • 发表日期：

    2026年3月31日

查看全文
  • Jinhee Park,
  • Hee Bin Yoo,
  • Byoung-Tak Zhang,
  • Junseok Kwon

摘要

This paper presents a novel perspective on enhancing the performance of classification-based Deep Metric Learning (DML). While classification-based DML has seen substantial progress through various regularization techniques, conventional normalization and scaling methods often lead to premature loss saturation and vanishing gradients caused by model overconfidence. To address this issue, we introduce Confidence Control (CC), a new regularization method that prevents this saturation by actively managing prediction confidence. By ensuring sufficient gradient magnitudes throughout training, CC encourages samples to align more strongly with their corresponding class weights. This results in improved feature invariance and tighter intra-class clustering in the embedding space. We propose two implementations of CC: (1) NaiveCC, a direct method that continuously maintains meaningful loss magnitudes by explicitly reducing confidence via a detached logit term; and (2) Evidence-based CC (EVDCC), which addresses the risk of excessive variance reduction (i.e., feature collapse) by imposing a geometric constraint based on eigenvector-based augmentation. Experimental results demonstrate that CC is easily integrable into existing DML frameworks and consistently improves performance, achieving a 2–5% gain in Recall@K.