论文标题
通过虚拟课程和示例开采改善深度度量学习
Improving Deep Metric Learning with Virtual Classes and Examples Mining
论文作者
论文摘要
在深度度量学习中，培训程序依赖于抽样信息元素。但是，随着训练程序的进行，如果没有适当的采矿策略或基于生成的方法，几乎不可能采样相关的硬负面示例。关于硬性负面生成的最新工作表现出了解决采矿问题的巨大承诺。但是，这一代过程很难调整，并且经常导致标记不正确的示例。为了解决这个问题，我们介绍了Mirage，这是一种基于一代的方法，该方法依赖于完全由生成的示例组成的虚拟类别，这些示例是培训类之间的缓冲区域。我们从经验上表明，与其他一代方法相比，虚拟类可显着改善流行数据集（CUB-200-2011，CARS-196和Stanford Online Products）的结果。
In deep metric learning, the training procedure relies on sampling informative tuples. However, as the training procedure progresses, it becomes nearly impossible to sample relevant hard negative examples without proper mining strategies or generation-based methods. Recent work on hard negative generation have shown great promises to solve the mining problem. However, this generation process is difficult to tune and often leads to incorrectly labelled examples. To tackle this issue, we introduce MIRAGE, a generation-based method that relies on virtual classes entirely composed of generated examples that act as buffer areas between the training classes. We empirically show that virtual classes significantly improve the results on popular datasets (Cub-200-2011, Cars-196 and Stanford Online Products) compared to other generation methods.
