Thomas Unterthiner

phone: +43-732-2468-4532
fax: +43-732-2468-4539

e-mail: unterthiner@bioinf...

Room: S3 327 (Computer Science building/Science Park 3)

Research Interests

Understanding the inner working of deep learning
Generative Adversarial Networks
Deep Learning and Neural Networks
Autonomous Driving
Drug Target Prediction
Machine Learning in Bioinformatics
Machine Learning and Artificial Intelligence

Selected Publications

Calvin Seward, Thomas Unterthiner, Urs Bergmann, Nikolay Jetchev, Sepp Hochreiter: First Order Generative Adversarial Networks, International Conference on Machine Learning (ICML), 2018, link
Andreas Mayr*, Günter Klambauer*, Thomas Unterthiner*, Marvin Steijaert, Jörg K. Wegner, Hugo Ceulemans, Djork-Arne Clevert, Sepp Hochreiter: Large-scale comparison of machine learning methods for drug target prediction on ChEMBL, Chemical Science, 2018, link
Thomas Unterthiner, Bernhard Nessler, Günter Klambauer, Martin Heusel, Hubert Ramsauer, Sepp Hochreiter: Coulomb GANs: Provably Optimal Nash Equilibria via Potential Fields, International Conference on Learning Representations (ICLR), 2018, link, code
Martin Heusel, Hubert Ramsauer, Thomas Unterthiner, Bernhard Nessler, Günter Klambauer, Sepp Hochreiter: GANs Trained by a Two Time-Scale Update Rule Converge to a Nash Equilibrium, Advances in Neural Information Processing Systems 30 (NIPS), 2017, link
Günter Klambauer, Thomas Unterthiner, Andreas Mayr, Sepp Hochreiter: Self-Normalizing Neural Networks, Advances in Neural Information Processing Systems 30 (NIPS), 2017, link
Djork-Arne Clevert, Thomas Unterthiner, Gundula Povysil, Sepp Hochreiter: Rectified factor networks for biclustering of omics data, Bioinformatics, 2017, link
Michael Treml*, José Arjona-Medina*, Thomas Unterthiner*, Rupesh Durgesh, Felix Friedmann, Peter Schuberth, Andreas Mayr, Martin Heusel, Markus Hofmarcher, Michael Widrich, Bernhard Nessler, Sepp Hochreiter Speeding up Semantic Segmentation for Autonomous Driving, Machine Learning for Intelligent Transportation Systems, in conjunction with Neural Information Processing Systems (NIPS), 2016, link
Thomas Unterthiner*, Andreas Mayr*, Günter Klambauer*, Sepp Hochreiter: Deep Learning for Drug Target Prediction, GPU Technology Conference Europe (GTC Europe), 2016,
Best Poster Award
Djork-Arne Clevert, Thomas Unterthiner, Sepp Hochreiter: Fast and Accurate Deep Network Learning by Exponential Linear Units (ELUs), International Conference on Learning Representations (ICLR), 2016, link
Andreas Mayr*, Günter Klambauer*, Thomas Unterthiner*, Sepp Hochreiter: DeepTox: Toxicity Prediction using Deep Learning, Frontiers in Environmental Science, 2016, link
Djork-Arne Clevert, Andreas Mayr, Thomas Unterthiner, Sepp Hochreiter: Rectified Factor Networks, Advances in Neural Information Processing Systems 28 (NIPS), 2015, link
Günter Klambauer, Martin Wischenbart, Michael Mahr, Thomas Unterthiner, Andreas Mayr, Sepp Hochreiter: Rchemcpp: a web service for structural analoging in ChEMBL, Drugbank and the Connectivity Map, Bioinformatics, 2015, link
Thomas Unterthiner, Andreas Mayr, Günter Klambauer, Sepp Hochreiter: Toxicity Prediction using Deep Learning, arXiv pre-print, 2015, link
Günter Klambauer, Bie Verbist, Liesbet Vervoort, Willem Talloen, The QSTAR Consortium, Ziv Shkedy, Olivier Thas, Andreas Bender, Hinrich W.H. Göhlmann, Sepp Hochreiter: Using transcriptomics to guide lead optimization in drug discovery projects: Lessons learned from the QSTAR project, Drug Discovery Today, 2015, link
Lev Givon, Thomas Unterthiner, N. Benjamin Erichson, David Wei Chiang, Eric Larson, Luke A. Pfister, Sander Dieleman, Gregory R. Lee, Stefan van der Walt, Teodor Mihai Moldovan, Frederic Bastien, Xing Shi, Jan Schlüter, Brian Thomas, Chris Capdevila, Alex Rubinsteyn: scikit-cuda 0.5.1: a Python interface to GPU-powered libraries, 2015, link
Aakash Chavan Ravindranath, Nolen Perualila-Tan, Adetayo Kasim, Georgios Drakakis, Sonia Liggi, Suzanne C. Brewerton, Daniel Mason, Michael J. Bodkin, David A. Evans, Aditya Bhagwat, Willem Talloen, Hinrich W. H. Göhlmann, QSTAR Consortium, Ziv Shkedy, Andreas Bender: Connecting gene expression data from connectivity map and in silico target predictions for small molecule mechanism-of-action analysis, Molecular BioSystems, 2014, link
Thomas Unterthiner, Andreas Mayr, Günter Klambauer, Marvin Steijaert, Jörg Wegner, Hugo Ceulemans, Sepp Hochreiter: Deep Learning as an Opportunity in Virtual Screening, Deep Learning and Representation Learning Workshop, in conjunction with Neural Information Processing Systems (NIPS), 2014, PDF
Günter Klambauer, Thomas Unterthiner, Sepp Hochreiter: DEXUS: Identifying Differential Expression in RNA-Seq Studies with Unknown Conditions, Nucleic Acids Research 41 (21), 2013, link
Thomas Unterthiner: Evaluating RNA-seq Methods for Differential Gene Expression, Master thesis, 2012
Thomas Unterthiner, Anne-Kathrin Schultz, Jan Bulla, Burkhard Morgenstern, Mario Stanke, Ingo Bulla: Detection of viral sequence fragments of HIV-1 subfamilies yet unknown. BMC Bioinformatics, 12(1):93, 2011, link

Teaching

2018 summer term: Deep Learning and Neural Networks
2017 summer term: Deep Learning and Neural Networks
2016 summer term: Deep Learning and Neural Networks
2015 summer term: Deep Learning and Neural Networks
2014 winter term: Exercises in Sequence Analysis and Phylogenetics
2014 summer term: Exercises in Theoretical Concepts of Machine Learning
2014 summer term: Exercises in Machine Learning: Unsupervised Techniques
2013 summer term: Exercises in Theoretical Bioinformatics and Machine Learning

Software

binet: A Deep Learning framework in Python with GPU support
librfn: high-performance implementation of Rectified Factor Networks, runs on CPU and GPU
libfabia: High performance parallel implementation of the FABIA biclustering algorithm
pyfabia: Cython implementation of FABIA
Regular contributor to various open source projects such as scikit-learn, PyCUDA, scikit-cuda and others

Education

2012 - present: PhD in Computer Science, Johannes Kepler University, Linz
2008-2016: MSc Computer Science (Diplom-Ingeneur), Johannes Kepler University, Linz, Austria (passed with distinction)
2008-2012: MSc Bioinformatics, Johannes Kepler University, Linz, Austria (passed with distinction)
2009-2010: Erasmus Study Abroad, University of Reading, United Kingdom
2005-2008: BSc Computer Science, Johannes Kepler University, Linz, Austria (passed with distinction)
2004: Matura (high school graduation) Deutschsprachige Handelsoberschule, Bolzano, Italy, Grade: 100/100

Employment Record

2012 - present: Research Assistant, Institute of Bioinformatics, Johannes Kepler University, Linz, Austria
2010 - 2011: Research Assistant, Department of Mathematics Education, Johannes Kepler University, Linz, Austria
2009: Research Assistant, Department of Bioinformatics, Institute for Microbiology and Genetics, University of Göttingen, Germany
2009: Summer-Internship as Software Developer at Orgachim ISC, Ruse, Bulgaria
2008: Summer-Internship as Software Developer at Würth-Phönix SRL, Bolzano, Italy