Get our free extension to see links to code for papers anywhere online!

Chrome logo Add to Chrome

Firefox logo Add to Firefox

Picture for Thomas Unterthiner

An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale


Oct 22, 2020
Alexey Dosovitskiy, Lucas Beyer, Alexander Kolesnikov, Dirk Weissenborn, Xiaohua Zhai, Thomas Unterthiner, Mostafa Dehghani, Matthias Minderer, Georg Heigold, Sylvain Gelly, Jakob Uszkoreit, Neil Houlsby

* Fine-tuning code and pre-trained models are available at https://github.com/google-research/vision_transformer 

  Access Paper or Ask Questions

Object-Centric Learning with Slot Attention


Jun 26, 2020
Francesco Locatello, Dirk Weissenborn, Thomas Unterthiner, Aravindh Mahendran, Georg Heigold, Jakob Uszkoreit, Alexey Dosovitskiy, Thomas Kipf


  Access Paper or Ask Questions

Predicting Neural Network Accuracy from Weights


Feb 26, 2020
Thomas Unterthiner, Daniel Keysers, Sylvain Gelly, Olivier Bousquet, Ilya Tolstikhin


  Access Paper or Ask Questions

Interpretable Deep Learning in Drug Discovery


Mar 18, 2019
Kristina Preuer, G√ľnter Klambauer, Friedrich Rippmann, Sepp Hochreiter, Thomas Unterthiner

* Code available at https://github.com/bioinf-jku/interpretable_ml_drug_discovery 

  Access Paper or Ask Questions

Towards Accurate Generative Models of Video: A New Metric & Challenges


Dec 03, 2018
Thomas Unterthiner, Sjoerd van Steenkiste, Karol Kurach, Raphael Marinier, Marcin Michalski, Sylvain Gelly


  Access Paper or Ask Questions

Fréchet ChemNet Distance: A metric for generative models for molecules in drug discovery


Aug 01, 2018
Kristina Preuer, Philipp Renz, Thomas Unterthiner, Sepp Hochreiter, G√ľnter Klambauer

* Implementations are available at: https://www.github.com/bioinf-jku/FCD 

  Access Paper or Ask Questions

RUDDER: Return Decomposition for Delayed Rewards


Jun 20, 2018
Jose A. Arjona-Medina, Michael Gillhofer, Michael Widrich, Thomas Unterthiner, Sepp Hochreiter

* 9 Pages plus appendix. For the code https://github.com/ml-jku/baselines-rudder. For videos https://goo.gl/EQerZV 

  Access Paper or Ask Questions

First Order Generative Adversarial Networks


Jun 07, 2018
Calvin Seward, Thomas Unterthiner, Urs Bergmann, Nikolay Jetchev, Sepp Hochreiter

* Accepted to 35th International Conference on Machine Learning (ICML). Code to reproduce experiments is available https://github.com/zalandoresearch/first_order_gan 

  Access Paper or Ask Questions

Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields


Jan 30, 2018
Thomas Unterthiner, Bernhard Nessler, Calvin Seward, G√ľnter Klambauer, Martin Heusel, Hubert Ramsauer, Sepp Hochreiter

* Published as a conference paper at ICLR (International Conference on Learning Representations) 2018. Implementation available at https://github.com/bioinf-jku/coulomb_gan 

  Access Paper or Ask Questions

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


Jan 12, 2018
Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, Sepp Hochreiter

* Advances in Neural Information Processing Systems 30 (NIPS 2017) 
* Implementations are available at: https://github.com/bioinf-jku/TTUR 

  Access Paper or Ask Questions

Self-Normalizing Neural Networks


Sep 07, 2017
G√ľnter Klambauer, Thomas Unterthiner, Andreas Mayr, Sepp Hochreiter

* Advances in Neural Information Processing Systems 30 (NIPS 2017) 
* 9 pages (+ 93 pages appendix) 

  Access Paper or Ask Questions

Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs)


Feb 22, 2016
Djork-Arné Clevert, Thomas Unterthiner, Sepp Hochreiter

* Published as a conference paper at ICLR 2016 

  Access Paper or Ask Questions

Rectified Factor Networks


Jun 11, 2015
Djork-Arné Clevert, Andreas Mayr, Thomas Unterthiner, Sepp Hochreiter

* Advances in Neural Information Processing Systems 28 (NIPS 2015) 
* 9 pages + 49 pages supplement 

  Access Paper or Ask Questions

Toxicity Prediction using Deep Learning


Mar 04, 2015
Thomas Unterthiner, Andreas Mayr, G√ľnter Klambauer, Sepp Hochreiter


  Access Paper or Ask Questions