Dual Adversarial Autoencoders for Clustering

• Published in: IEEE transaction on neural networks and learning systems Vol. 31; no. 4; pp. 1417 - 1424
• Main Authors: Ge, Pengfei, Ren, Chuan-Xian, Dai, Dao-Qing, Feng, Jiashi, Yan, Shuicheng
• Format: Journal Article
• Language: English
• Published: United States IEEE 01.04.2020; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: AAE; Algorithms; Benchmarks; Clustering; Clustering algorithms; Clustering methods; Computer vision; Data analysis; deep generative models; Gallium nitride; Generative adversarial networks; Image reconstruction; Information processing; latent variable; Mutual information; mutual information regularization; Objective function; Pattern recognition; Regularization; Task analysis; Training
• ISSN: 2162-237X; 2162-2388
• DOI: 10.1109/TNNLS.2019.2919948

As a powerful approach for exploratory data analysis, unsupervised clustering is a fundamental task in computer vision and pattern recognition. Many clustering algorithms have been developed, but most of them perform unsatisfactorily on the data with complex structures. Recently, adversarial autoencoder (AE) (AAE) shows effectiveness on tackling such data by combining AE and adversarial training, but it cannot effectively extract classification information from the unlabeled data. In this brief, we propose dual AAE (Dual-AAE) which simultaneously maximizes the likelihood function and mutual information between observed examples and a subset of latent variables. By performing variational inference on the objective function of Dual-AAE, we derive a new reconstruction loss which can be optimized by training a pair of AEs. Moreover, to avoid mode collapse, we introduce the clustering regularization term for the category variable. Experiments on four benchmarks show that Dual-AAE achieves superior performance over state-of-the-art clustering methods. In addition, by adding a reject option, the clustering accuracy of Dual-AAE can reach that of supervised CNN algorithms. Dual-AAE can also be used for disentangling style and content of images without using supervised information.