Learning Disentangled Representations via Mutual Information Estimation

2019/12/09 by Eduardo Hugo Sanchez, Sanchez, Eduardo Hugo, Mathieu Serrurier +3 · 4 citations
Computer Science · #Adversarial Robustness in Machine Learning #Digital Media Forensic Detection #FOS: Computer and information sciences #Generative Adversarial Networks and Image Synthesis #Machine Learning (cs.LG) #Machine Learning (stat.ML)

paper · doi:10.48550/arxiv.1912.03915

openalex publication_date 2019/12/09 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

In this paper, we investigate the problem of learning disentangled representations. Given a pair of images sharing some attributes, we aim to create a low-dimensional representation which is split into two parts: a shared representation that captures the common information between the images and an exclusive representation that contains the specific information of each image. To address this issue, we propose a model based on mutual information estimation without relying on image reconstruction or image generation. Mutual information maximization is performed to capture the attributes of data in the shared and exclusive representations while we minimize the mutual information between the shared and exclusive representation to enforce representation disentanglement. We show that these representations are useful to perform downstream tasks such as image classification and image retrieval based on the shared or exclusive component. Moreover, classification results show that our model outperforms the state-of-the-art model based on VAE/GAN approaches in representation disentanglement.

Citations

• Auto-Encoding Variational Bayes
• Learning deep representations by mutual information estimation and maximization
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

Cited by

• GED-Consistent Disentanglement of Aligned and Unaligned Substructures for Graph Similarity Learning
• DiA-gnostic VLVAE: Disentangled Alignment-Constrained Vision Language Variational AutoEncoder for Robust Radiology Reporting with Missing Modalities
• NMINE: Normalized Mutual Information Neural Estimation
• 2D Gaussians Meet Visual Tokenizer

Related

• Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation
• On Mutual Information Maximization for Representation Learning
• HSIC-InfoGAN: Learning Unsupervised Disentangled Representations by Maximising Approximated Mutual Information
• Towards Better Understanding of Disentangled Representations via Mutual Information
• Disentangled Representation Learning with Information Maximizing Autoencoder