OOGAN: Disentangling GAN with One-Hot Sampling and Orthogonal Regularization

Exploring the potential of GANs for unsupervised disentanglement learning, this paper proposes a novel GAN-based disentanglement framework with One-Hot Sampling and Orthogonal Regularization (OOGAN). While previous works mostly attempt to tackle disentanglement learning through VAE and seek to implicitly minimize the Total Correlation (TC) objective with various sorts of approximation methods, we show that GANs have a natural advantage in disentangling with an alternating latent variable (noise) sampling method that is straightforward and robust. Furthermore, we provide a brand-new perspective on designing the structure of the generator and discriminator, demonstrating that a minor structural change and an orthogonal regularization on model weights entails an improved disentanglement. Instead of experimenting on simple toy datasets, we conduct experiments on higher-resolution images and show that OOGAN greatly pushes the boundary of unsupervised disentanglement.

• Publication year

  2019

• Venue

  AAAI Conference on Artificial Intelligence

• Publication date

  2019-05-26

• Fields of study

  Mathematics, Computer Science

• Identifiers
  DOI 10.1609/AAAI.V34I04.5919arXiv 1905.10836
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• AUTO-ENCODING VARIATIONAL BAYES

  2020influential reference
• Semantic-Guided Multi-Attention Localization for Zero-Shot Learning

  2019cited by this paper
• Learning Feature-to-Feature Translator by Alternating Back-Propagation for Generative Zero-Shot Learning

  2019cited by this paper
• InfoGAN-CR: Disentangling Generative Adversarial Networks with Contrastive Regularizers

  2019cited by this paper
• Learning Feature-to-Feature Translator by Alternating Back-Propagation for Zero-Shot Learning

  2019cited by this paper
• On the information bottleneck theory of deep learning

  2018cited by this paper
• Self-Attention Generative Adversarial Networks

  2018influential reference
• Understanding disentangling in β-VAE

  2018cited by this paper
• IB-GAN: Disentangled Representation Learning with Information Bottleneck GAN

  2018influential reference
• Can We Gain More from Orthogonality Regularizations in Training Deep CNNs?

  2018cited by this paper
• GENERATIVE ADVERSARIAL NETS

  2018cited by this paper
• Spectral Normalization for Generative Adversarial Networks

  2018cited by this paper
• A Style-Based Generator Architecture for Generative Adversarial Networks

  2018cited by this paper
• Structured Disentangled Representations

  2018cited by this paper
• Challenging Common Assumptions in the Unsupervised Learning of Disentangled Representations

  2018cited by this paper
• Joint-VAE: Learning Disentangled Joint Continuous and Discrete Representations

  2018cited by this paper
• A Framework for the Quantitative Evaluation of Disentangled Representations

  2018cited by this paper
• The Shape of Art History in the Eyes of the Machine

  2018cited by this paper
• Disentangling by Factorising

  2018influential reference
• Isolating Sources of Disentanglement in Variational Autoencoders

  2018cited by this paper
• Finding Principal Semantics of Style in Art

  2018cited by this paper
• Entropy and mutual information in models of deep neural networks

  2018cited by this paper
• Orthogonal Weight Normalization: Solution to Optimization over Multiple Dependent Stiefel Manifolds in Deep Neural Networks

  2017cited by this paper
• Progressive Growing of GANs for Improved Quality, Stability, and Variation

  2017cited by this paper
• Bayesian Compression for Deep Learning

  2017cited by this paper
• Link the Head to the "Beak": Zero Shot Learning from Noisy Text Description at Part Precision

  2017cited by this paper
• ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices

  2017cited by this paper
• CAN: Creative Adversarial Networks, Generating "Art" by Learning About Styles and Deviating from Style Norms

  2017cited by this paper
• Fader Networks: Manipulating Images by Sliding Attributes

  2017cited by this paper
• An Information-Theoretic Analysis of Deep Latent-Variable Models

  2017cited by this paper
• A Generative Adversarial Approach for Zero-Shot Learning from Noisy Texts

  2017cited by this paper
• Information Dropout: Learning Optimal Representations Through Noisy Computation

  2016cited by this paper
• InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

  2016influential reference
• What can the brain teach us about building artificial intelligence?

  2016cited by this paper
• Neural Photo Editing with Introspective Adversarial Networks

  2016cited by this paper
• Categorical Reparameterization with Gumbel-Softmax

  2016cited by this paper
• beta-VAE: Learning Basic Visual Concepts with a Constrained Variational Framework

  2016influential reference
• Amortised MAP Inference for Image Super-resolution

  2016cited by this paper
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks

  2015cited by this paper
• Deep Learning Face Attributes in the Wild

  2014cited by this paper
• Very Deep Convolutional Networks for Large-Scale Image Recognition

  2014cited by this paper
• Seeing 3D Chairs: Exemplar Part-Based 2D-3D Alignment Using a Large Dataset of CAD Models

  2014cited by this paper
• ImageNet classification with deep convolutional neural networks

  2012cited by this paper
• Representation Learning: A Review and New Perspectives

  2012cited by this paper
• A Mathematical Theory of Communication

  2006cited by this paper
• Information Theoretical Analysis of Multivariate Correlation

  1960cited by this paper

• Interpretable Visual Neural Decoding with Unsupervised Semantic Disentanglement

  2025cites this paper
• Sparsification and Reconstruction from the Perspective of Representation Geometry

  2025cites this paper
• Disentangled representation learning for multi-label generalized zero-shot learning

  2025cites this paper
• Towards Robust and Reliable Concept Representations: Reliability-Enhanced Concept Embedding Model

  2025cites this paper
• Unsupervised Latent Regression through Information Maximization - Contrastive Regularized GAN

  2024cites this paper
• Unsupervised Panoptic Interpretation of Latent Spaces in GANs Using Space-Filling Vector Quantization

  2024cites this paper
• Efficient Zero-Knowledge Arguments For Paillier Cryptosystem

  2024cites this paper
• Human internal state estimation as blind source separation using a dynamic auto-encoder

  2023cites this paper
• PC-IEN: a click-through rate prediction method based on dynamic collaborative personalized interest extraction

  2023cites this paper
• Generated Image Editing Method Based on Global-Local Jacobi Disentanglement for Machine Learning

  2023cites this paper
• ProtoVAE: Prototypical Networks for Unsupervised Disentanglement

  2023cites this paper
• Consistency Preservation and Feature Entropy Regularization for GAN Based Face Editing

  2023cites this paper
• medXGAN: Visual Explanations for Medical Classifiers through a Generative Latent Space

  2022cites this paper
• EAD-GAN: A Generative Adversarial Network for Disentangling Affine Transforms in Images

  2022cites this paper
• DOT-VAE: Disentangling One Factor at a Time

  2022cites this paper
• DISENTANGLED GENERATIVE MODELS AND THEIR APPLICATIONS

  2022cites this paper
• Deep Curvilinear Editing: Commutative and Nonlinear Image Manipulation for Pretrained Deep Generative Model

  2022cites this paper
• Facial Attribute Editing based on Independent Selective Transfer Unit and Self-attention Mechanism

  2022cites this paper
• WaterPurifier: A scalable system to prevent the DNS water torture attack in 5G-enabled SIoT network

  2022cites this paper
• Training and Evaluation of Deep Policies using Reinforcement Learning and Generative Models

  2022cites this paper
• Generalized Zero-Shot Learning using Identifiable Variational Autoencoders

  2021cites this paper
• Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis

  2021cites this paper
• Convolutional Normalization: Improving Deep Convolutional Network Robustness and Training

  2021cites this paper
• CoDeGAN: Contrastive Disentanglement for Generative Adversarial Network

  2021cites this paper
• EigenGAN: Layer-Wise Eigen-Learning for GANs

  2021cites this paper
• MichiGAN: sampling from disentangled representations of single-cell data using generative adversarial networks

  2021cites this paper
• A Multifeature Learning and Fusion Network for Facial Age Estimation

  2021cites this paper
• A Tutorial on Learning Disentangled Representations in the Imaging Domain

  2021cites this paper
• WarpedGANSpace: Finding non-linear RBF paths in GAN latent space

  2021cites this paper
• Inference-InfoGAN: Inference Independence via Embedding Orthogonal Basis Expansion

  2021cites this paper
• End-to-End Speech Separation Using Orthogonal Representation in Complex and Real Time-Frequency Domain

  2021cites this paper
• Learning disentangled representations in the imaging domain

  2021cites this paper
• Delta-GAN-Encoder: Encoding Semantic Changes for Explicit Image Editing, using Few Synthetic Samples

  2021cites this paper
• Cluster-guided Image Synthesis with Unconditional Models

  2021cites this paper
• ABSent: Cross-Lingual Sentence Representation Mapping with Bidirectional GANs

  2020cites this paper
• Normalization Techniques in Training DNNs: Methodology, Analysis and Application

  2020cites this paper
• Data-efficient visuomotor policy training using reinforcement learning and generative models

  2020cites this paper
• Unsupervised Controllable Generation with Self-Training

  2020cites this paper
• Deep Isometric Learning for Visual Recognition

  2020cites this paper
• High-Fidelity Synthesis with Disentangled Representation

  2020cites this paper
• 2D Image Relighting with Image-to-Image Translation

  2020cites this paper
• Unsupervised Discovery of Interpretable Directions in the GAN Latent Space

  2020cites this paper
• InfoGAN-CR and ModelCentrality: Self-supervised Model Training and Selection for Disentangling GANs

  2019cites this paper