arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1801.01587stat.MLcs.LG
收藏

SpectralNet:使用深度神经网络的谱聚类

SpectralNet: Spectral Clustering using Deep Neural Networks

Yale University(耶鲁大学);Weizmann Institute of Science(魏茨曼科学研究所)

Uri Shaham, Kelly Stanton, Henry Li, Boaz Nadler, Ronen Basri, Yuval Kluger

AI总结:

本文提出SpectralNet,一种基于深度神经网络的谱聚类方法,通过约束随机优化学习嵌入,克服可扩展性和泛化问题,并在Reuters数据集上取得最先进聚类结果。

摘要:

谱聚类是无监督数据分析中领先且流行的技术。它的两个主要局限是可扩展性以及谱嵌入的泛化(即样本外扩展)。在本文中,我们引入了一种深度学习方法来解决谱聚类中的上述缺点。我们的网络,称为SpectralNet,学习一个映射,将输入数据点嵌入到其关联的图拉普拉斯矩阵的特征空间中,随后对它们进行聚类。我们使用涉及约束随机优化的过程来训练SpectralNet。随机优化使其能够扩展到大型数据集,而约束(通过专用输出层实现)使我们能够保持网络输出的正交性。此外,SpectralNet学习的映射自然地将谱嵌入泛化到未见过的数据点。为了进一步提高聚类质量,我们使用Siamese网络从无标签数据中学习亲和度,替代标准的成对高斯亲和度。通过将网络应用于标准自编码器产生的编码表示,可以实现额外的改进。我们的端到端学习过程是完全无监督的。此外,我们应用VC维理论推导出SpectralNet规模的下界。在Reuters数据集上报告了最先进的聚类结果。我们的实现可在https://github.com/kstant0725/SpectralNet公开获取。

英文摘要:

Spectral clustering is a leading and popular technique in unsupervised data analysis. Two of its major limitations are scalability and generalization of the spectral embedding (i.e., out-of-sample-extension). In this paper we introduce a deep learning approach to spectral clustering that overcomes the above shortcomings. Our network, which we call SpectralNet, learns a map that embeds input data points into the eigenspace of their associated graph Laplacian matrix and subsequently clusters them. We train SpectralNet using a procedure that involves constrained stochastic optimization. Stochastic optimization allows it to scale to large datasets, while the constraints, which are implemented using a special-purpose output layer, allow us to keep the network output orthogonal. Moreover, the map learned by SpectralNet naturally generalizes the spectral embedding to unseen data points. To further improve the quality of the clustering, we replace the standard pairwise Gaussian affinities with affinities leaned from unlabeled data using a Siamese network. Additional improvement can be achieved by applying the network to code representations produced, e.g., by standard autoencoders. Our end-to-end learning procedure is fully unsupervised. In addition, we apply VC dimension theory to derive a lower bound on the size of SpectralNet. State-of-the-art clustering results are reported on the Reuters dataset. Our implementation is publicly available at https://github.com/kstant0725/SpectralNet .

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。

↑