Gaussian Mixture Generative Adversarial Networks for Diverse Datasets, and the Unsupervised Clustering of Images

Abstract

Generative Adversarial Networks (GANs) have been shown to produce realistically looking synthetic images with remarkable success, yet their performance seems less impressive when the training set is highly diverse. In order to provide a better fit to the target data distribution when the dataset includes many different classes, we propose a variant of the basic GAN model, called Gaussian Mixture GAN (GM-GAN), where the probability distribution over the latent space is a mixture of Gaussians. We also propose a supervised variant which is capable of conditional sample synthesis. In order to evaluate the model's performance, we propose a new scoring method which separately takes into account two (typically conflicting) measures - diversity vs. quality of the generated data. Through a series of empirical experiments, using both synthetic and real-world datasets, we quantitatively show that GM-GANs outperform baselines, both when evaluated using the commonly used Inception Score, and when evaluated using our own alternative scoring method. In addition, we qualitatively demonstrate how the \textit{unsupervised} variant of GM-GAN tends to map latent vectors sampled from different Gaussians in the latent space to samples of different classes in the data space. We show how this phenomenon can be exploited for the task of unsupervised clustering, and provide quantitative evaluation showing the superiority of our method for the unsupervised clustering of image datasets. Finally, we demonstrate a feature which further sets our model apart from other GAN models: the option to control the quality-diversity trade-off by altering, post-training, the probability distribution of the latent space. This allows one to sample higher quality and lower diversity samples, or vice versa, according to one's needs.

20 pages, 8 figures

References in corpus (4)

• Learning Discrete Representations via Information Maximizing Self-Augmented Training
• Deep Clustering via Joint Convolutional Autoencoder Embedding and Relative Entropy Minimization
• Clustering and Unsupervised Anomaly Detection with L2 Normalized Deep Auto-Encoder Representations
• Learning Latent Representations in Neural Networks for Clustering through Pseudo Supervision and Graph-based Activity Regularization

Cited by in corpus (11)

• Generalization Properties of Optimal Transport GANs with Latent Distribution Learning
• MMGAN: Generative Adversarial Networks for Multi-Modal Distributions
• Max-Affine Spline Insights into Deep Generative Networks
• Unsupervised multi-modal Styled Content Generation
• GMM-Based Generative Adversarial Encoder Learning
• MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining
• Non-Exhaustive Learning Using Gaussian Mixture Generative Adversarial Networks
• Disentangling Latent Space for VAE by Label Relevant/Irrelevant Dimensions
• Stylized Text Generation Using Wasserstein Autoencoders with a Mixture of Gaussian Prior
• Multiclass non-Adversarial Image Synthesis, with Application to Classification from Very Small Sample
• Label-Removed Generative Adversarial Networks Incorporating with K-Means