从间隔（Margins）视角理解对比学习
Understanding Contrastive Learning Through the Lens of Margins
AI总结：
该研究从梯度分析新视角拆解间隔对对比学习梯度的影响，明确强化正样本、按正样本角度与logits缩放梯度是提升其泛化性能的关键，为对比学习改进提供方向。
摘要：
对比学习及其变体是在多领域均表现优异的自监督学习方法。对比学习采用余弦相似度衡量表征间的距离，并使用交叉熵损失进行表征学习。在同样基于余弦相似度的表征学习框架下，间隔（margins）已在提升人脸识别与说话人识别任务性能中发挥了重要作用。有趣的是，尽管二者依赖相同的相似度度量与目标函数，对比学习尚未主动引入间隔机制。此外，目前仅存在基于决策边界的解释用于说明间隔在对比学习中的作用。本研究提出了一种基于梯度分析的新视角，用于理解间隔的作用。基于该新视角，我们分析了间隔如何影响对比学习的梯度，并将其作用拆解为更基础的层面。我们对各层面分别展开分析，并提出了改进对比学习的潜在方向。实验结果表明，强化正样本作用、根据正样本角度与logits对梯度进行缩放，是提升对比学习在已见和未见数据集上泛化性能的关键，其余因素仅能带来小幅性能提升。
英文摘要：
Contrastive learning, along with its variations, has been a highly effective self-supervised learning method across diverse domains. Contrastive learning measures the distance between representations using cosine similarity and uses cross-entropy for representation learning. Within the same framework of cosine-similarity-based representation learning, margins have played a significant role in enhancing face and speaker recognition tasks. Interestingly, despite the shared reliance on the same similarity metrics and objective functions, contrastive learning has not actively adopted margins. Furthermore, decision-boundary-based explanations are the only ones that have been used to explain the effect of margins in contrastive learning. In this work, we propose a new perspective to understand the role of margins based on gradient analysis. Based on the new perspective, we analyze how margins affect gradients of contrastive learning and separate the effect into more elemental levels. We separately analyze each and provide possible directions for improving contrastive learning. Our experimental results demonstrate that emphasizing positive samples and scaling gradients depending on positive sample angles and logits are the keys to improving the generalization performance of contrastive learning in both seen and unseen datasets, and other factors can only marginally improve performance.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
