论文标题
负余量问题：了解少量分类的利润
Negative Margin Matters: Understanding Margin in Few-shot Classification
论文作者
论文摘要
本文引入了基于公制学习的几杆学习方法的负差损失。负边缘损失的表现明显优于常规的软马克斯损失，并且在三个标准的几杆分类基准的基准上，铃铛和哨声很少。这些结果与公制学习领域中的共同实践背道而驰，即边缘为零或阳性。为了理解为什么负差损失在几次分类的情况下表现良好，我们分析了学到的特征W.R.T的可区分性W.R.T的不同边缘在经验和理论上，用于培训和新颖的课程。我们发现，尽管负边缘降低了训练类别的特征可区分性，但它也可能避免将同一新颖类的错误映射到多个峰或群集，从而使新阶级的歧视有益。代码可在https://github.com/bl0/negation-margin.few-shot上找到。
This paper introduces a negative margin loss to metric learning based few-shot learning methods. The negative margin loss significantly outperforms regular softmax loss, and achieves state-of-the-art accuracy on three standard few-shot classification benchmarks with few bells and whistles. These results are contrary to the common practice in the metric learning field, that the margin is zero or positive. To understand why the negative margin loss performs well for the few-shot classification, we analyze the discriminability of learned features w.r.t different margins for training and novel classes, both empirically and theoretically. We find that although negative margin reduces the feature discriminability for training classes, it may also avoid falsely mapping samples of the same novel class to multiple peaks or clusters, and thus benefit the discrimination of novel classes. Code is available at https://github.com/bl0/negative-margin.few-shot.
