Bernhard Nessler — accepted AI-conference papers

2021

The balancing principle for parameter choice in distance-regularized domain adaptation

Werner Zellinger, Natalia Shepeleva, Marius-Constantin Dinu, Hamid Eghbal-zadeh, Hoan Duc Nguyen, Bernhard Nessler +2

NeurIPS 2021poster

We address the unsolved algorithm design problem of choosing a justified regularization parameter in unsupervised domain adaptation. This problem is intriguing as no labels are available in the target domain. Our approach starts with the observation that the widely-used method of minimizing the sour…

Cited by 10 Source ↗PDF ↗code ↗Scholar ↗

2018

Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields

Thomas Unterthiner, Bernhard Nessler, Calvin Seward, Günter Klambauer, Martin Heusel, Hubert Ramsauer +1

ICLR 2018poster

Generative adversarial networks (GANs) evolved into one of the most successful unsupervised techniques for generating realistic images. Even though it has recently been shown that GAN training converges, GAN models often end up in local Nash equilibria that are associated with mode collapse or other…

Cited by 84 Source ↗PDF ↗code ↗Scholar ↗

2017

GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium

Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, Sepp Hochreiter

NeurIPS 2017poster

Generative Adversarial Networks (GANs) excel at creating realistic images with complex models for which maximum likelihood is infeasible. However, the convergence of GAN training has still not been proved. We propose a two time-scale update rule (TTUR) for training GANs with stochastic gradient desc…

Cited by 17120 Source ↗PDF ↗code ↗Scholar ↗