Least Squares Generative Adversarial Networks

Abstract

Unsupervised learning with generative adversarial networks (GANs) has proven hugely successful. Regular GANs hypothesize the discriminator as a classifier with the sigmoid cross entropy loss function. However, we found that this loss function may lead to the vanishing gradients problem during the learning process. To overcome such a problem, we propose in this paper the Least Squares Generative Adversarial Networks (LSGANs) which adopt the least squares loss function for the discriminator. We show that minimizing the objective function of LSGAN yields minimizing the Pearson ${\chi }^2$ divergence. There are two benefits of LSGANs over regular GANs. First, LSGANs are able to generate higher quality images than regular GANs. Second, LSGANs perform more stable during the learning process. We evaluate LSGANs on five scene datasets and the experimental results show that the images generated by LSGANs are of better quality than the ones generated by regular GANs. We also conduct two comparison experiments between LSGANs and regular GANs to illustrate the stability of LSGANs.

References in corpus (16)

• Adam: A Method for Stochastic Optimization
• Auto-Encoding Variational Bayes
• Conditional Generative Adversarial Nets
• Unsupervised Representation Learning with Deep Convolutional Generative Adversarial Networks
• Generative Adversarial Text to Image Synthesis
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• Improved Techniques for Training GANs
• LSUN: Construction of a Large-scale Image Dataset using Deep Learning with Humans in the Loop
• InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets
• Energy-based Generative Adversarial Network
• Estimating divergence functionals and the likelihood ratio by convex risk minimization
• f-GAN: Training Generative Neural Samplers using Variational Divergence Minimization
• Mode Regularized Generative Adversarial Networks
• Learning in Implicit Generative Models
• Plug & Play Generative Networks: Conditional Iterative Generation of Images in Latent Space
• Loss-Sensitive Generative Adversarial Networks on Lipschitz Densities

Cited by in corpus (91)

• Progressive Growing of GANs for Improved Quality, Stability, and Variation
• Improved Training of Wasserstein GANs
• Reconstruction of three-dimensional porous media using generative adversarial neural networks
• How Generative Adversarial Networks and Their Variants Work: An Overview
• Adversarial Learning for Semi-Supervised Semantic Segmentation
• Sharpness-aware Low dose CT denoising using conditional generative adversarial network
• An empirical study on evaluation metrics of generative adversarial networks
• Generative Image Inpainting with Contextual Attention
• Parallel-Data-Free Voice Conversion Using Cycle-Consistent Adversarial Networks
• Automatic Goal Generation for Reinforcement Learning Agents
• Deep Generative Adversarial Networks for Compressed Sensing Automates MRI
• SPG-Net: Segmentation Prediction and Guidance Network for Image Inpainting
• Boundary-Seeking Generative Adversarial Networks
• Adversarial Generation of Natural Language
• Generative Models for Automatic Chemical Design
• Data Augmentation in Emotion Classification Using Generative Adversarial Networks
• Future Frame Prediction for Anomaly Detection -- A New Baseline
• On the Effects of Batch and Weight Normalization in Generative Adversarial Networks
• Semantic-aware Grad-GAN for Virtual-to-Real Urban Scene Adaption
• Robust Imitation of Diverse Behaviors
• Geometry Score: A Method For Comparing Generative Adversarial Networks
• A Multi-Pass GAN for Fluid Flow Super-Resolution
• One-Sided Unsupervised Domain Mapping
• Photographic Text-to-Image Synthesis with a Hierarchically-nested Adversarial Network
• Learning Independent Features with Adversarial Nets for Non-linear ICA
• Generative adversarial network-based glottal waveform model for statistical parametric speech synthesis
• Activation Maximization Generative Adversarial Nets
• ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing
• Recurrent Generative Adversarial Networks for Proximal Learning and Automated Compressive Image Recovery
• Understanding GANs: the LQG Setting
• Visually-Aware Fashion Recommendation and Design with Generative Image Models
• Text to Image Synthesis Using Generative Adversarial Networks
• Adversarial Sparse-View CBCT Artifact Reduction
• Class-Splitting Generative Adversarial Networks
• SPIGAN: Privileged Adversarial Learning from Simulation
• Training Shallow and Thin Networks for Acceleration via Knowledge Distillation with Conditional Adversarial Networks
• Image to Image Translation for Domain Adaptation
• GAN-QP: A Novel GAN Framework without Gradient Vanishing and Lipschitz Constraint
• TextureGAN: Controlling Deep Image Synthesis with Texture Patches
• Unsupervised Cipher Cracking Using Discrete GANs
• Gang of GANs: Generative Adversarial Networks with Maximum Margin Ranking
• Image Clustering using an Augmented Generative Adversarial Network and Information Maximization
• All about Structure: Adapting Structural Information across Domains for Boosting Semantic Segmentation
• Towards Realistic 3D Embedding via View Alignment
• Graph Generative Adversarial Networks for Sparse Data Generation in High Energy Physics
• Unsupervised Person Image Synthesis in Arbitrary Poses
• Triple consistency loss for pairing distributions in GAN-based face synthesis
• CANZSL: Cycle-Consistent Adversarial Networks for Zero-Shot Learning from Natural Language
• AlignGAN: Learning to Align Cross-Domain Images with Conditional Generative Adversarial Networks
• From source to target and back: symmetric bi-directional adaptive GAN
• Cross-domain Human Parsing via Adversarial Feature and Label Adaptation
• TequilaGAN: How to easily identify GAN samples
• Pulsar Candidate Identification with Artificial Intelligence Techniques
• Composite Functional Gradient Learning of Generative Adversarial Models
• Likelihood Estimation for Generative Adversarial Networks
• EncryptGAN: Image Steganography with Domain Transform
• Understanding the Effectiveness of Lipschitz-Continuity in Generative Adversarial Nets
• Mix and match networks: encoder-decoder alignment for zero-pair image translation
• RankGAN: A Maximum Margin Ranking GAN for Generating Faces
• Decoupled Learning for Conditional Adversarial Networks
• Metric Learning-based Generative Adversarial Network
• Conditional Adversarial Generative Flow for Controllable Image Synthesis
• Generative adversarial networks for generation and classification of physical rehabilitation movement episodes
• Normalized Wasserstein Distance for Mixture Distributions with Applications in Adversarial Learning and Domain Adaptation
• Learning Implicit Generative Models with the Method of Learned Moments
• Variance Regularizing Adversarial Learning
• Cycle-Consistent Adversarial Learning as Approximate Bayesian Inference
• Probabilistic Generative Adversarial Networks
• Smart, Sparse Contours to Represent and Edit Images
• Ebola Optimization Search Algorithm (EOSA): A new metaheuristic algorithm based on the propagation model of Ebola virus disease
• Correcting differences in multi-site neuroimaging data using Generative Adversarial Networks
• DivCo: Diverse Conditional Image Synthesis via Contrastive Generative Adversarial Network
• Gotta Adapt 'Em All: Joint Pixel and Feature-Level Domain Adaptation for Recognition in the Wild
• Learning Disentangled Representations of Satellite Image Time Series
• FaceShapeGene: A Disentangled Shape Representation for Flexible Face Image Editing
• Butterfly Effect: Bidirectional Control of Classification Performance by Small Additive Perturbation
• Facial Feature Embedded CycleGAN for VIS-NIR Translation
• AutoRemover: Automatic Object Removal for Autonomous Driving Videos
• On the Use of Audio Fingerprinting Features for Speech Enhancement with Generative Adversarial Network
• PBGen: Partial Binarization of Deconvolution-Based Generators for Edge Intelligence
• Exploiting Images for Video Recognition with Hierarchical Generative Adversarial Networks
• Towards Real-Time Advancement of Underwater Visual Quality with GAN
• Single-channel Speech Dereverberation via Generative Adversarial Training
• Synthetic Dynamic PMU Data Generation: A Generative Adversarial Network Approach
• Local Facial Makeup Transfer via Disentangled Representation
• Implicit Subspace Prior Learning for Dual-Blind Face Restoration
• Analysis of Macula on Color Fundus Images Using Heightmap Reconstruction Through Deep Learning
• Towards GANs' Approximation Ability
• MURAUER: Mapping Unlabeled Real Data for Label AUstERity
• Latent Transformations for Object View Points Synthesis
• Selective Sampling and Mixture Models in Generative Adversarial Networks