Abstract
Distance-based classification is frequently used in transductive few-shot learning (FSL). However, due to the high-dimensionality of image representations, FSL classifiers are prone to suffer from the hubness problem, where a few points (hubs) occur frequently in multiple nearest neighbour lists of other points. Hubness negatively impacts distance-based classification when hubs from one class appear often among the nearest neighbors of points from another class, degrading the classifier's performance. To address the hubness problem in FSL, we first prove that hubness can be eliminated by distributing representations uniformly on the hypersphere. We then propose two new approaches to embed representations on the hypersphere, which we prove optimize a tradeoff between uniformity and local similarity preservation -- reducing hubness while retaining class structure. Our experiments show that the proposed methods reduce hubness, and significantly improves transductive FSL accuracy for a wide range of classifiers.
Abstract (translated)
基于距离的分类常常用于递归 few-shot 学习(FSL)。然而,由于图像表示的高维度性,FSL 分类器容易受到中心化问题的困扰。其中一些点(中心点)经常出现在其他点的所有近邻列表中的多个点中。中心化在一类点中的中心点出现在另一类点中的近邻点中时会消极影响基于距离的分类,从而降低分类器的性能。为了解决 FSL 中的中心化问题,我们首先证明可以通过均匀分布表示在球面上消除中心化。然后我们提出了两个新的方法和在球面上嵌入表示的方法,我们证明可以优化均匀性和局部相似保留之间的权衡,以减少中心化同时保留类别结构。我们的实验结果表明,提出的方法和对各类别的分类器显著改善了递归 FSL 的准确性。
URL
https://arxiv.org/abs/2303.09352