arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1809.04157cs.LGcs.CVstat.ML
收藏

加热Softmax嵌入

Heated-Up Softmax Embedding

Columbia University(哥伦比亚大学);Google Research(谷歌研究院)

Xu Zhang, Felix Xinnan Yu, Svebor Karaman, Wei Zhang, Shih-Fu Chang

摘要:

度量学习的目标是学习一个与样本语义一致的距离。这个问题通常通过学习每个样本的嵌入来解决,使得相同类别的样本的嵌入是紧凑的,而不同类别的样本的嵌入在特征空间中展开。我们研究了从基于深度神经网络的分类器的倒数第二层提取的特征,该分类器在softmax层的顶部使用交叉熵损失进行训练。我们表明,使用不同温度值的softmax函数训练分类器会导致具有不同紧凑度的特征。利用这些见解,我们提出了一个“加热”的策略,训练一个分类器,随着温度的升高,导致相应的嵌入,以实现国家的最先进的性能在各种度量学习基准。

英文摘要:

Metric learning aims at learning a distance which is consistent with the semantic meaning of the samples. The problem is generally solved by learning an embedding for each sample such that the embeddings of samples of the same category are compact while the embeddings of samples of different categories are spread-out in the feature space. We study the features extracted from the second last layer of a deep neural network based classifier trained with the cross entropy loss on top of the softmax layer. We show that training classifiers with different temperature values of softmax function leads to features with different levels of compactness. Leveraging these insights, we propose a "heating-up" strategy to train a classifier with increasing temperatures, leading the corresponding embeddings to achieve state-of-the-art performance on a variety of metric learning benchmarks.

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。

↑