Energy-based Generative Adversarial Network

Abstract

We introduce the "Energy-based Generative Adversarial Network" model (EBGAN) which views the discriminator as an energy function that attributes low energies to the regions near the data manifold and higher energies to other regions. Similar to the probabilistic GANs, a generator is seen as being trained to produce contrastive samples with minimal energies, while the discriminator is trained to assign high energies to these generated samples. Viewing the discriminator as an energy function allows to use a wide variety of architectures and loss functionals in addition to the usual binary classifier with logistic output. Among them, we show one instantiation of EBGAN framework as using an auto-encoder architecture, with the energy being the reconstruction error, in place of the discriminator. We show that this form of EBGAN exhibits more stable behavior than regular GANs during training. We also show that a single-scale architecture can be trained to generate high-resolution images.

Submitted to ICLR 2017

References in corpus (1)

• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks

Cited by in corpus (54)

• CyCADA: Cycle-Consistent Adversarial Domain Adaptation
• Mode Regularized Generative Adversarial Networks
• A Connection between Generative Adversarial Networks, Inverse Reinforcement Learning, and Energy-Based Models
• Learning in Implicit Generative Models
• Adversarial Feature Matching for Text Generation
• Age Progression/Regression by Conditional Adversarial Autoencoder
• Connecting Generative Adversarial Networks and Actor-Critic Methods
• On the Origin of Deep Learning
• Best of Both Worlds: Transferring Knowledge from Discriminative Learning to a Generative Visual Dialog Model
• Learning to Draw Samples: With Application to Amortized MLE for Generative Adversarial Learning
• Triangle Generative Adversarial Networks
• Adversarial Network Embedding
• MAGAN: Margin Adaptation for Generative Adversarial Networks
• Multi-Generator Generative Adversarial Nets
• Adversarial Symmetric Variational Autoencoder
• Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields
• Modeling documents with Generative Adversarial Networks
• High-Quality Face Image SR Using Conditional Generative Adversarial Networks
• eCommerceGAN : A Generative Adversarial Network for E-commerce
• An Adversarial Regularisation for Semi-Supervised Training of Structured Output Neural Networks
• Generative Adversarial Networks as Variational Training of Energy Based Models
• GANGs: Generative Adversarial Network Games
• Adversarial Network Bottleneck Features for Noise Robust Speaker Verification
• Symmetric Variational Autoencoder and Connections to Adversarial Learning
• Learning Deep Energy Models: Contrastive Divergence vs. Amortized MLE
• DeLiGAN : Generative Adversarial Networks for Diverse and Limited Data
• Task Specific Adversarial Cost Function
• Distributional Adversarial Networks
• Ways of Conditioning Generative Adversarial Networks
• Recent Progress of Face Image Synthesis
• Deep Co-Space: Sample Mining Across Feature Transformation for Semi-Supervised Learning
• Proximal Alternating Direction Network: A Globally Converged Deep Unrolling Framework
• Differential Generative Adversarial Networks: Synthesizing Non-linear Facial Variations with Limited Number of Training Data
• Annealed Generative Adversarial Networks
• TripletGAN: Training Generative Model with Triplet Loss
• Energy-Based Sequence GANs for Recommendation and Their Connection to Imitation Learning
• Face Translation between Images and Videos using Identity-aware CycleGAN
• Metric Learning-based Generative Adversarial Network
• Deep Stacked Networks with Residual Polishing for Image Inpainting
• Improving Direct Physical Properties Prediction of Heterogeneous Materials from Imaging Data via Convolutional Neural Network and a Morphology-Aware Generative Model
• Acoustic Feature Learning via Deep Variational Canonical Correlation Analysis
• Probabilistic Generative Adversarial Networks
• Image Quality Assessment Techniques Show Improved Training and Evaluation of Autoencoder Generative Adversarial Networks
• Can GAN Learn Topological Features of a Graph?
• Generative Adversarial Mapping Networks
• Pipeline Generative Adversarial Networks for Facial Images Generation with Multiple Attributes
• Towards Grounding Conceptual Spaces in Neural Representations
• Depth Structure Preserving Scene Image Generation
• Semi-Supervised Learning via New Deep Network Inversion
• Visual Data Augmentation through Learning
• On reproduction of On the regularization of Wasserstein GANs
• Generative Adversarial Networks with Inverse Transformation Unit
• Linear Discriminant Generative Adversarial Networks
• Instance Map based Image Synthesis with a Denoising Generative Adversarial Network