paper-with-me

홈 › Papers

On gradient regularizers for MMD GANs

2018-05-29 · NeurIPS 2018 12 · Michael Arbel, Danica J. Sutherland, Mikołaj Bińkowski, Arthur Gretton

We propose a principled method for gradient-based regularization of the critic of GAN-like models trained by adversarially optimizing the kernel of a Maximum Mean Discrepancy (MMD). We show that controlling the gradient of the critic is vital to having a sensible loss function, and devise a method to enforce exact, analytical gradient constraints at no additional cost compared to existing approximate techniques based on additive regularizers. The new loss function is provably continuous, and experiments show that it stabilizes and accelerates training, giving image generation models that outperform state-of-the art methods on $160 \times 160$ CelebA and $64 \times 64$ unconditional ImageNet.

📄 PDF Abstract BibTeX arXiv:1805.11565

Code (1)

MichaelArbel/Scaled-MMD-GAN 공식 구현 tf

Tasks

Image Generation

Similar Papers 제목 키워드 기반

Quality Aware Generative Adversarial Networks

2019-11-08 · NeurIPS 2019 12 · Parimala Kancharla, Sumohana S. Channappayya

Generative Adversarial Networks (GANs) have become a very popular tool for implicitly learning high-dimensional probability distributions. Several improvements have been made to the original GAN formulation to address so…

Full reference image quality assessmentFull-Reference Image Quality AssessmentImage GenerationImage Quality Assessment+2

Two steps at a time --- taking GAN training in stride with Tseng's method

2021-01-01 · Axel Böhm, Michael Sedlmayer, Ernö Robert Csetnek, Radu Ioan Bot

Motivated by the training of Generative Adversarial Networks (GANs), we study methods for solving minimax problems with additional nonsmooth regularizers. We do so by employing \emph{monotone operator} theory, in particu…

Wasserstein Proximal of GANs

2021-02-13 · ICLR 2019 5 · Alex Tong Lin, Wuchen Li, Stanley Osher, Guido Montufar

We introduce a new method for training generative adversarial networks by applying the Wasserstein-2 metric proximal on the generators. The approach is based on Wasserstein information geometry. It defines a parametrizat…

GANs with Variational Entropy Regularizers: Applications in Mitigating the Mode-Collapse Issue

2020-09-24 · Pirazh Khorramshahi, Hossein Souri, Rama Chellappa, Soheil Feizi

Building on the success of deep learning, Generative Adversarial Networks (GANs) provide a modern approach to learn a probability distribution from observed samples. GANs are often formulated as a zero-sum game between t…

Diversity

A General Family of Stochastic Proximal Gradient Methods for Deep Learning

2020-07-15 · Jihun Yun, Aurelie C. Lozano, Eunho Yang

We study the training of regularized neural networks where the regularizer can be non-smooth and non-convex. We propose a unified framework for stochastic proximal gradient descent, which we term ProxGen, that allows for…

Quantization