GAN Loss

GAN loss functions are crucial components of Generative Adversarial Networks (GANs), determining how the generator and discriminator interact during training to produce high-quality synthetic data. Current research focuses on improving GAN training stability and the quality of generated samples by exploring alternative loss functions, such as those based on class probability estimation or f-divergences, and integrating them with other loss types like vector quantization losses. These advancements aim to address issues like mode collapse, blurry outputs, and training instability, ultimately leading to more robust and efficient GAN models with applications spanning image synthesis, 3D modeling, and anomaly detection.

Papers

December 8, 2024

Accelerating Video Diffusion Models via Distribution Matching
Yuanzhi Zhu, Hanshu Yan, Huan Yang, Kai Zhang, Junnan Li
Generative Model Video Generation Video Diffusion Model Distribution Matching Diffusion Distillation Pre Trained Image Diffusion Model GAN Loss

May 23, 2024

Improved Distribution Matching Distillation for Fast Image Synthesis
Tianwei Yin, Michaël Gharbi, Taesung Park, Richard Zhang, Eli Shechtman, Fredo Durand, William T. Freeman
Image Generation Image Synthesis GAN Loss Training Inference Distribution Matching Distillation

December 19, 2023

Self-supervised Learning for Enhancing Geometrical Modeling in 3D-Aware Generative Adversarial Network
Jiarong Guo, Xiaogang Xu, Hengshuang Zhao
Generative Adversarial Network Self Supervised Learning Geometric Modeling GAN Loss Constrained Generative

October 27, 2023

Addressing GAN Training Instabilities via Tunable Classification Losses
Monica Welfert, Gowtham R. Kurri, Kyle Otstot, Lalitha Sankar
GAN Training GAN Loss F GAN

August 9, 2023

Vector quantization loss analysis in VQGANs: a single-GPU ablation study for image-to-image synthesis
Luv Verma, Varun Mohan
Image Synthesis Quantization Loss GAN Loss Vector Quantized Generative Adversarial Network

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

May 12, 2022

$\alpha$-GAN: Convergence and Estimation Guarantees
Gowtham R. Kurri, Monica Welfert, Tyler Sypherd, Lalitha Sankar
Early Stage Convergence GAN Loss GAN Framework

March 20, 2022

RareGAN: Generating Samples for Rare Classes
Zinan Lin, Hao Liang, Giulia Fanti, Vyas Sekar
Generative Adversarial Network Sample Generation GAN Loss Rare Class

February 2, 2022

Robust Estimation for Nonparametric Families via Generative Adversarial Networks
Banghua Zhu, Jiantao Jiao, Michael I. Jordan
Generative Adversarial Network Robust Estimation GAN Loss High Dimensional Robust Kolmogorov Smirnov

January 26, 2022

Generative Trees: Adversarial and Copycat
Richard Nock, Mathieu Guillame-Bert
Generative Adversarial Network Feature Imitation Adversarial Text GAN Loss

November 12, 2021

Contrastive Feature Loss for Image Prediction
Alex Andonian, Taesung Park, Bryan Russell, Phillip Isola, Jun-Yan Zhu, Richard Zhang
GAN Model Contrastive Loss Image Synthesis High Similarity Image Prediction GAN Loss

November 4, 2021

GraN-GAN: Piecewise Gradient Normalization for Generative Adversarial Networks
Vineeth S. Bhaskara, Tristan Aumentado-Armstrong, Allan Jepson, Alex Levinshtein
Generative Adversarial Network Feature Normalization GAN Loss Gradient Normalization