Unsupervised Deep Learning by Neighbourhood Discovery
Jiabo Huang Qi Dong Shaogang Gong Xiatian Zhu
Abstract
Deep convolutional neural networks (CNNs) have demonstrated remarkable success in computer vision by supervisedly learning strong visual feature representations. However, training CNNs relies heavily on the availability of exhaustive training data annotations, limiting significantly their deployment and scalability in many application scenarios. In this work, we introduce a generic unsupervised deep learning approach to training deep models without the need for any manual label supervision. Specifically, we progressively discover sample anchored/centred neighbourhoods to reason and learn the underlying class decision boundaries iteratively and accumulatively. Every single neighbourhood is specially formulated so that all the member samples can share the same unseen class labels at high probability for facilitating the extraction of class discriminative feature representations during training. Experiments on image classification show the performance advantages of the proposed method over the state-of-the-art unsupervised learning models on six benchmarks including both coarse-grained and fine-grained object image categorisation.
中文速览
深度卷积神经网络在图像识别上表现出色，但训练时需要海量人工标注数据，现实中往往难以满足。为此，研究者提出了一种名为"锚邻域发现"（Anchor Neighbourhood Discovery，AND）的无监督深度学习方法，核心思路是：既不像聚类那样强行把所有样本划入全局类别（容易出错），也不像样本特异性学习那样把每个样本都当成独立类（忽略样本间关系），而是以每个训练样本为中心，在特征空间中动态发现一小群高度相似的"邻域"，并假设同一邻域内的样本属于同一类，以此为监督信号逐步训练模型、优化特征；同时引入课程学习策略，随着模型不断改善而逐渐扩大邻域、提升邻域的类别一致性。在六个图像分类基准（包括粗粒度的CIFAR10/100、SVHN、ImageNet和细粒度的CUB200、Stanford Dogs）上的实验表明，AND方法全面优于当时的无监督学习前沿模型。这项工作的意义在于，它为不依赖任何人工标注的通用视觉特征学习提供了一条可扩展的新路径。
原文 arXiv:1904.11567;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1904.11567v3