论文标题
无监督表示学习的统一混合物观看框架
A Unified Mixture-View Framework for Unsupervised Representation Learning
论文作者
论文摘要
最近的无监督对比表示学习遵循单个实例多视图（SIM）范式，其中通常用图像内图数据增强构建正对。在本文中，我们提出了一种有效的方法，称为超越单个实例多视图（BSIM）。具体而言，我们通过测量两个随机抽样实例及其混合物（即伪阳性对）之间的关节相似性来施加更准确的实例歧视能力。我们认为，学习联合相似性有助于提高编码功能在潜在空间中均匀分布时的性能。我们将其用作无监督的对比表示学习的正交改进，包括当前的杰出方法Simclr，Moco和Byol。我们在许多下游基准测试中评估了我们的学会表示，例如Imagenet-1K和Pascal VOC 2007的线性分类，Coco 2017和VOC上的对象检测等。与先前的艺术相比，我们在所有这些任务上获得了很大的利润，我们获得了很大的增长。
Recent unsupervised contrastive representation learning follows a Single Instance Multi-view (SIM) paradigm where positive pairs are usually constructed with intra-image data augmentation. In this paper, we propose an effective approach called Beyond Single Instance Multi-view (BSIM). Specifically, we impose more accurate instance discrimination capability by measuring the joint similarity between two randomly sampled instances and their mixture, namely spurious-positive pairs. We believe that learning joint similarity helps to improve the performance when encoded features are distributed more evenly in the latent space. We apply it as an orthogonal improvement for unsupervised contrastive representation learning, including current outstanding methods SimCLR, MoCo, and BYOL. We evaluate our learned representations on many downstream benchmarks like linear classification on ImageNet-1k and PASCAL VOC 2007, object detection on MS COCO 2017 and VOC, etc. We obtain substantial gains with a large margin almost on all these tasks compared with prior arts.
