GAN Training

GAN training aims to stabilize and improve the performance of Generative Adversarial Networks, which generate realistic data by pitting a generator against a discriminator in a minimax game. Current research focuses on novel loss functions, regularization techniques (e.g., Lipschitz continuity constraints, score matching), and architectural modifications to address issues like mode collapse, overfitting (especially with limited data), and training instability. These advancements are significant because they enhance the quality and diversity of generated data, impacting diverse fields such as medical imaging, video generation, and image editing, where high-fidelity synthetic data is crucial for various applications.

Papers

May 12, 2023

Spider GAN: Leveraging Friendly Neighbors to Accelerate GAN Training
Siddarth Asokan, Chandra Sekhar Seelamantula
Generative Adversarial Network Pre Trained GAN Training Thy Neighbor GAN Algorithm

May 7, 2023

SynthMix: Mixing up Aligned Synthesis for Medical Cross-Modality Domain Adaptation
Xinwen Zhang, Chaoyi Zhang, Dongnan Liu, Qianbi Yu, Weidong Cai
Domain Adaptation GAN Model Synthetic Image Critical Synthesis GAN Training Segmentation Benchmark Adversarial Method

May 4, 2023

Multimodal-driven Talking Face Generation via a Unified Diffusion-based Generator
Chao Xu, Shaoting Zhu, Junwei Zhu, Tianxin Huang, Jiangning Zhang, Ying Tai, Yong Liu
Diffusion Model Multimodal Phenomenon Unified Alignment Face Generation Diffusion Based GAN Training Texture Transfer

April 29, 2023

LD-GAN: Low-Dimensional Generative Adversarial Network for Spectral Image Generation with Variance Regularization
Emmanuel Martinez, Roman Jacome, Alejandra Hernandez-Rojas, Henry Arguello
GAN Model GAN Training Variance Regularization

March 27, 2023

Sequential training of GANs against GAN-classifiers reveals correlated "knowledge gaps" present among independently trained GAN instances
Arkanath Pathak, Nicholas Dufour
Generative Adversarial Network Sequential Learning GAN Training Knowledge Gap

March 24, 2023

A Three-Player GAN for Super-Resolution in Magnetic Resonance Imaging
Qi Wang, Lucas Mahler, Julius Steiglechner, Florian Birk, Klaus Scheffler, Gabriele Lohmann
Generative Adversarial Network Super Resolution Magnetic Resonance Imaging GAN Training Magnetic Resonance Wasserstein Gan GAN Loss

March 13, 2023

SDF-3DGAN: A 3D Object Generative Method Based on Implicit Signed Distance Function
Lutao Jiang, Ruyi Ji, Libo Zhang
3D Generative GAN Training Signed Distance Function 3D Object Generation

March 2, 2023

Improving GAN Training via Feature Space Shrinkage
Haozhe Liu, Wentian Zhang, Bing Li, Haoqian Wu, Nanjun He, Yawen Huang, Yuexiang Li, Bernard Ghanem, Yefeng Zheng
Generative Adversarial Network GAN Training State of the Art GAN Feature Reduction

February 28, 2023

February 8, 2023

Domain Adaptation of Synthetic Driving Datasets for Real-World Autonomous Driving
Koustav Mullick, Harshil Jain, Sanchit Gupta, Amit Arvind Kale
Domain Adaptation Synthetic Data GAN Training Domain Translation Real World Autonomous Domain Adaptation Task Synthetic Driving

February 6, 2023

Private GANs, Revisited
Alex Bie, Gautam Kamath, Guojun Zhang
GAN Training Representation Drift Discriminative Training Private Generative Adversarial Network

January 30, 2023

SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer
Yuhta Takida, Masaaki Imaizumi, Takashi Shibuya, Chieh-Hsin Lai, Toshimitsu Uesaka, Naoki Murata, Yuki Mitsufuji
Generative Adversarial Network GAN Model Adversarial Learning GAN Training Discriminative Reply Linear Layer Approximability Hierarchy

January 11, 2023

LinkGAN: Linking GAN Latents to Pixels for Controllable Image Synthesis
Jiapeng Zhu, Ceyuan Yang, Yujun Shen, Zifan Shi, Bo Dai, Deli Zhao, Qifeng Chen
Pixel Level Tetromino Pixel GAN Inversion GAN Training Latent Code GAN Generator Controllable Image Synthesis

December 8, 2022

Effective Dynamics of Generative Adversarial Networks
Steven Durr, Youssef Mroueh, Yuhai Tu, Shenshen Wang
Generative Adversarial Network Adversarial Learning Generative Question GAN Training Effective Dynamic

November 27, 2022

DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data
Tiantian Fang, Ruoyu Sun, Alex Schwing
GAN Model Limited Data Loss Landscape GAN Training

November 8, 2022

Does an ensemble of GANs lead to better performance when training segmentation networks with synthetic images?
Måns Larsson, Muhammad Usman Akbar, Anders Eklund
GAN Model Synthetic Image Diverse Ensemble Segmentation Network Performance Improvement GAN Training Single GAN

October 29, 2022

Recursive Reasoning in Minimax Games: A Level $k$ Gradient Play Method
Zichu Liu, Lacra Pavel
Generative Adversarial Network Practical Method Gradient Norm GAN Training Different Level Convergence Behavior Recursive Thinking Minimax Game

October 28, 2022

Latent Space is Feature Space: Regularization Term for GANs Training on Limited Dataset
Pengwei Wang
Generative Adversarial Network Latent Space GAN Model Limited Data Feature Space GAN Training GAN Based Regularization Term

October 27, 2022

ScoreMix: A Scalable Augmentation Strategy for Training GANs with Limited Data
Jie Cao, Mandi Luo, Junchi Yu, Ming-Hsuan Yang, Ran He
Generative Adversarial Network Limited Data GAN Training Effective Augmentation Score Estimation Network

GAN Training

Papers

Spider GAN: Leveraging Friendly Neighbors to Accelerate GAN Training

SynthMix: Mixing up Aligned Synthesis for Medical Cross-Modality Domain Adaptation

Multimodal-driven Talking Face Generation via a Unified Diffusion-based Generator

LD-GAN: Low-Dimensional Generative Adversarial Network for Spectral Image Generation with Variance Regularization

Sequential training of GANs against GAN-classifiers reveals correlated "knowledge gaps" present among independently trained GAN instances

A Three-Player GAN for Super-Resolution in Magnetic Resonance Imaging

SDF-3DGAN: A 3D Object Generative Method Based on Implicit Signed Distance Function

Improving GAN Training via Feature Space Shrinkage

Balanced Training for Sparse GANs

$(\alpha_D,\alpha_G)$-GANs: Addressing GAN Training Instabilities via Dual Objectives

Domain Adaptation of Synthetic Driving Datasets for Real-World Autonomous Driving

Private GANs, Revisited

SAN: Inducing Metrizability of GAN with Discriminative Normalized Linear Layer

LinkGAN: Linking GAN Latents to Pixels for Controllable Image Synthesis

Effective Dynamics of Generative Adversarial Networks

DigGAN: Discriminator gradIent Gap Regularization for GAN Training with Limited Data

Does an ensemble of GANs lead to better performance when training segmentation networks with synthetic images?

Recursive Reasoning in Minimax Games: A Level $k$ Gradient Play Method

Latent Space is Feature Space: Regularization Term for GANs Training on Limited Dataset

ScoreMix: A Scalable Augmentation Strategy for Training GANs with Limited Data