Elastic-InfoGAN: Unsupervised Disentangled Representation Learning in Class-Imbalanced Data

• DOI: 10.48550/arxiv.1910.01112
• Publication Date: 2019-01-01
• Authors: Ojha, Utkarsh; Singh, Krishna Kumar; Hsieh, Cho-Jui; Lee, Yong Jae
• Keywords: FOS: Computer and information sciences; Computer Science - Machine Learning; 03 medical and health sciences; 0302 clinical medicine; Statistics - Machine Learning; Computer Vision and Pattern Recognition (cs.CV); Computer Science - Computer Vision and Pattern Recognition; Machine Learning (stat.ML); Machine Learning (cs.LG)

Camera ready version for NeurIPS 2020<br/>We propose a novel unsupervised generative model that learns to disentangle object identity from other low-level aspects in class-imbalanced data. We first investigate the issues surrounding the assumptions about uniformity made by InfoGAN, and demonstrate its ineffectiveness to properly disentangle object identity in imbalanced data. Our key idea is to make the discovery of the discrete latent factor of variation invariant to identity-preserving transformations in real images, and use that as a signal to learn the appropriate latent distribution representing object identity. Experiments on both artificial (MNIST, 3D cars, 3D chairs, ShapeNet) and real-world (YouTube-Faces) imbalanced datasets demonstrate the effectiveness of our method in disentangling object identity as a latent factor of variation.<br/>