论文标题
proxynca ++：重新审视和振兴代理社区组件分析
ProxyNCA++: Revisiting and Revitalizing Proxy Neighborhood Component Analysis
论文作者
论文摘要
我们考虑距离度量学习（DML）的问题，其中任务是学习图像之间的有效相似性度量。我们重新访问proxynca并结合了几种增强功能。我们发现低温缩放是一种至关重要的成分，并解释了它的作用原因。此外，我们还发现，与全球平均池相比，全球最大池量通常效果更好。此外，我们提出的快速移动代理还解决了代理的小梯度问题，并且该组成部分与低温缩放和全局最大池相当协同。与原始的Proxynca算法相比，我们的增强模型（称为Proxynca ++）在四个不同的零射击检索数据集中取得了22.9个百分点的平均召回率提高。此外，我们在Cub200，CARS196，SOP和Inshop数据集上实现了最新的结果，分别达到72.2、90.1、81.4和90.9的召回率。
We consider the problem of distance metric learning (DML), where the task is to learn an effective similarity measure between images. We revisit ProxyNCA and incorporate several enhancements. We find that low temperature scaling is a performance-critical component and explain why it works. Besides, we also discover that Global Max Pooling works better in general when compared to Global Average Pooling. Additionally, our proposed fast moving proxies also addresses small gradient issue of proxies, and this component synergizes well with low temperature scaling and Global Max Pooling. Our enhanced model, called ProxyNCA++, achieves a 22.9 percentage point average improvement of Recall@1 across four different zero-shot retrieval datasets compared to the original ProxyNCA algorithm. Furthermore, we achieve state-of-the-art results on the CUB200, Cars196, Sop, and InShop datasets, achieving Recall@1 scores of 72.2, 90.1, 81.4, and 90.9, respectively.
