Siamese Prototypical Contrastive Learning-猫眼课题宝

DOI：

10.48550/arxiv.2208.08819

发表时间：

2022-08

期刊：

ArXiv

影响因子：

通讯作者：

中科院分区：

其他

文献类型：

作者：

Shentong Mo;Zhun Sun;Chao Li

对比自我监督学习(CSL)是一种实用的解决方案，它以一种无监督的方式从海量数据中学习有意义的视觉表征。普通CSL将从神经网络中提取的特征嵌入到特定的拓扑结构中。在训练过程中，对比损失将同一输入的不同观点集中在一起，同时将来自不同输入的嵌入分开。CSL的缺点之一是损失项需要大量的负样本来提供理想的更好的互信息界。然而，通过更大的运行批处理大小增加负样本的数量也会增强假否定的效果：语义上相似的样本被推离锚点，从而降低了下游性能。在本文中，我们通过引入一个简单而有效的对比学习框架来解决这个问题。关键的见解是使用暹罗风格的度量损失来匹配原型内特征，同时增加原型间特征之间的距离。我们在不同的基准上进行了大量的实验，实验结果证明了我们的方法在提高视觉表示质量方面的有效性。具体地说，我们的无监督预训练ResNet-50具有线性探头，在ImageNet-1K数据集上的表现优于完全监督的训练版本。

Contrastive Self-supervised Learning (CSL) is a practical solution that learns meaningful visual representations from massive data in an unsupervised approach. The ordinary CSL embeds the features extracted from neural networks onto specific topological structures. During the training progress, the contrastive loss draws the different views of the same input together while pushing the embeddings from different inputs apart. One of the drawbacks of CSL is that the loss term requires a large number of negative samples to provide better mutual information bound ideally. However, increasing the number of negative samples by larger running batch size also enhances the effects of false negatives: semantically similar samples are pushed apart from the anchor, hence downgrading downstream performance. In this paper, we tackle this problem by introducing a simple but effective contrastive learning framework. The key insight is to employ siamese-style metric loss to match intra-prototype features, while increasing the distance between inter-prototype features. We conduct extensive experiments on various benchmarks where the results demonstrate the effectiveness of our method on improving the quality of visual representations. Specifically, our unsupervised pre-trained ResNet-50 with a linear probe, out-performs the fully-supervised trained version on the ImageNet-1K dataset.

Siamese Prototypical Contrastive Learning

文献摘要

相似文献