ClusterGAN : Latent Space Clustering in Generative Adversarial Networks

By Sudipto Mukherjee, Himanshu Asnani, Eugene Lin, Sreeram Kannan. AAAI Conference on Artificial Intelligence, 2018

384 citations recorded by the source corpus.

Generative Adversarial networks (GANs) have obtained remarkable success in many unsupervised learning tasks and unarguably, clustering is an important unsupervised learning problem. While one can potentially exploit the latent-space back-projection in GANs to cluster, we demonstrate that the cluster structure is not retained in the GAN latent space. In this paper, we propose ClusterGAN as a new mechanism for clustering using GANs. By sampling latent variables from a mixture of one-hot encoded variables and continuous latent variables, coupled with an inverse network (which projects the data to the latent space) trained jointly with a clustering specific loss, we are able to achieve clustering in the latent space. Our results show a remarkable phenomenon that GANs can preserve latent space interpolation across categories, even though the discriminator is never exposed to such vectors. We compare our results with various clustering baselines and demonstrate superior performance on both synthetic and real datasets.

Subfields: Computer Vision and Pattern Recognition, Signal Processing

View Paper: https://www.semanticscholar.org/paper/0943ba71d5a4d839a4d2d88a8aee72025fb77982

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

In Signal Processing

Citations in the corpus

Unsupervised multi-modal Styled Content Generation (2020, background): “2016], or ClusterGAN [Mukherjee et al. 2019], which combines the advantages of VAEs and GANs.”
Diverse Image Generation via Self-Conditioned GANs (2020, methodology): “A recent method [32] proposes to obtain good clustering using GANs, while we aim to achieve realistic and diverse generation.”
Generative Adversarial Networks (GANs) (2020, methodology): “Feature matching is similar to techniques that use maximum mean discrepancy [138]–[140] for G’s network training [128][129].”
Deep Spectral Clustering Using Dual Autoencoder Network (2019, methodology): “…deep embedded regularized clustering (DEPICT) [10], improved deep embedded clustering with locality preservation (IDEC) [12], deep spectral clustering with a set of nearest neighbor pairs (SpectralNet) [36], clustering with GAN (ClusterGAN) [30] and GAN with the mutual information (InfoGAN) [5].”