动态调整相似度学习中的间距参数，让模型持续挑战难样本。
Maintaining Difficulty: A Margin Scheduler for Triplet Loss in Siamese Networks Training
- 根据每轮难样本比例自动调节间距大小
- 在4个数据集上验证准确率均优于固定或递增间距
- 适合需要稳定优化距离度量的对比学习场景
三元组间隔排序损失是解决距离度量学习问题中最常用的损失函数之一，依赖于一个定义正负样本对最小距离的间隔参数μ。本文发现，在训练过程中，许多三元组的有效间隔往往超过预设的μ值，前提是观察到足够多违反该间隔的三元组。这一现象表明，固定间隔可能限制学习过程。基于此，我们提出一种间隔调度器，根据每轮观察到的易样本比例动态调整μ值，以保持训练难度恒定。实验结果表明，该策略在四个不同数据集上均优于固定间隔和单调递增间隔方案，显著提升了验证性能。
原文摘要 · Abstract (English)
The Triplet Margin Ranking Loss is one of the most widely used loss functions in Siamese Networks for solving Distance Metric Learning (DML) problems. This loss function depends on a margin parameter μ, which defines the minimum distance that should separate positive and negative pairs during training. In this work, we show that, during training, the effective margin of many triplets often exceeds the predefined value of μ, provided that a sufficient number of triplets violating this margin is observed. This behavior indicates that fixing the margin throughout training may limit the learning process. Based on this observation, we propose a margin scheduler that adjusts the value of μ according to the proportion of easy triplets observed at each epoch, with the goal of maintaining training difficulty over time. We show that the proposed strategy leads to improved performance when compared to both a constant margin and a monotonically increasing margin scheme. Experimental results on four different datasets show consistent gains in verification performance.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。