Simon Kohl
Research Scientist at DeepMind
I recently graduated with a PhD in Computer Science from KIT & DKFZ. In my thesis I developed probabilistic segmentation networks that allow to model distributions over plausible segmentations in the presence of ambiguity.
More broadly, I am interested in uncertainty handling, robustness and multi-modal outputs in deep (often generative) networks and am excited to push their boundaries in medical imaging and structural biology.
Publications
Contrastive Training for Improved Out-of-Distribution Detection
Jim Winkens, Rudy Bunel, Abhijit Guha Roy, Robert Stanforth, Vivek Natarajan, Joseph R. Ledsam, Patricia MacWilliams, Pushmeet Kohli, Alan Karthikesalingam, Simon Kohl, Taylan Cemgil, S. M. Ali Eslami, Olaf Ronneberger
arXiv 2020 (under review)
pdf
arXiv 2020
A Hierarchical Probabilistic U-Net for Modeling Multi-Scale Ambiguities
Simon A. A. Kohl, Bernardino Romera-Paredes, Klaus H. Maier-Hein, Danilo Jimenez Rezende, S. M. Ali Eslami, Pushmeet Kohli, Andrew Zisserman, Olaf Ronneberger
Medical Imaging Workshop, NeurIPS 2019
pdf code data+weights oral
Med-NeurIPS 2019
Classification of Cancer at Prostate MRI: Deep Learning versus Clinical PI-RADS Assessment
Patrick Schelb, Simon Kohl, Jan Philipp Radtke, Manuel Wiesenfarth, Philipp Kickingereder, Sebastian Bickelhaupt, Tristan Anselm Kuder, Albrecht Stenzinger, Markus Hohenfellner, Heinz-Peter Schlemmer, Klaus H Maier-Hein and David Bonekamp
RSNA Radiology 2019
pdf code editorial
Radiology 2019
Deep Probabilistic Modeling of Glioma Growth
Jens Petersen, Paul F. Jäger, Fabian Isensee, Simon A. A. Kohl, Ulf Neuberger, Wolfgang Wick, Jürgen Debus, Sabine Heiland, Martin Bendszus, Philipp Kickingereder, Klaus H. Maier-Hein
MICCAI 2019
pdf code
MICCAI 2019
Unsupervised Anomaly Localization using Variational Auto-Encoders
David Zimmerer, Fabian Isensee, Jens Petersen, Simon Kohl, Klaus Maier-Hein
MICCAI 2019
pdf
MICCAI 2019
Automated design of deep learning methods for biomedical image segmentation
Fabian Isensee, Jens Petersen, Simon A. A. Kohl, Paul F. Jäger, Klaus H. Maier-Hein
arXiv 2019
pdf code
arXiv 2019
Retina U-Net: Embarrassingly Simple Exploitation of Segmentation Supervision for Medical Object Detection
Paul F Jaeger, Simon AA Kohl, Sebastian Bickelhaupt, Fabian Isensee, Tristan Anselm Kuder, Heinz-Peter Schlemmer, Klaus H Maier-Hein
Machine Learning for Health Workshop, NeurIPS 2019
pdf code
ML4H Workshop
NeurIPS 2019
A Probabilistic U-Net for Segmentation of Ambiguous Images
Simon AA Kohl, Bernardino Romera-Paredes, Clemens Meyer, Jeffrey De Fauw, Joseph R Ledsam, Klaus H Maier-Hein, SM Ali Eslami, Danilo Jimenez Rezende, and Olaf Ronneberger
Neural Information Processing Systems (NeurIPS'18)
pdf spotlight video tf code pytorch re-code
NeurIPS 2018
A Case for the Score: Identifying Image Anomalies using Variational Autoencoder Gradients
David Zimmerer, Jens Petersen, Simon AA Kohl, and Klaus Maier-Hein
Medical Imaging Workshop at NeurIPS 2018
pdf workshop
Med-NeurIPS 2018
nnU-Net: Self-adapting Framework for U-Net-Based Medical Image Segmentation
Fabian Isensee, Jens Petersen, Andre Klein, David Zimmerer, Paul F Jaeger, Simon Kohl, Jakob Wasserthal, Gregor Koehler, Tobias Norajitra, Sebastian Wirkert, and Klaus H Maier-Hein
MICCAI 2018 Challenge: Medical Segmentation Decathlon (winning entry)
pdf leaderboard code
MICCAI 2018
Radiomic Machine Learning for Characterization of Prostate Lesions with MRI: Comparison to ADC Values
David Bonekamp*, Simon Kohl*, Manuel Wiesenfarth, Patrick Schelb, Jan Philipp Radtke, Michael Götz, Philipp Kickingereder, Kaneschka Yaqubi, Bertram Hitthaler, Nils Gählert, Tristan Anselm Kuder, Fenja Deister, Martin Freitag, Markus Hohenfellner, Boris A Hadaschik, Heinz-Peter Schlemmer, and Klaus H Maier-Hein
RSNA Radiology 2018
pdf editorial
Radiology 2018
Adversarial Networks for the Detection of Aggressive Prostate Cancer
Simon Kohl, David Bonekamp, Heinz-Peter Schlemmer, Kaneschka Yaqubi, Markus Hohenfellner, Boris Hadaschik, Jan-Philipp Radtke, and Klaus Maier-Hein
ML4H Workshop at Neural Information Processing Systems 2017
pdf pdf (long) workshop
ML4H Workshop
NeurIPS 2017
Invited Talks
A Hierachical Probabilistic U-Net for Modeling Multi-Scale Ambiguities
Oral Presentation @ ML4H NeurIPS Workshop, Vancouver, December 2019
recording
A Probabilistic U-Net for Segmentation of Ambiguous Images
Düsseldorf Bio-DataSeminar, University of Düsseldorf, September 2018
Visual Learning Lab, University of Heidelberg, October 2018
NEC Laboratories Europe, Heidelberg, November 2018
Mini-NeurIPS, CSML Seminar Series, University College London, November 2018
Spotlight Presentation @ NeurIPS, Montreal, December 2018
Oxford Applied and Theoretical Machine Learning Group, University of Oxford, February 2019
slides
From Unstructured Data to Medically Relevant Information
ISMRM Nordic Chapter Meeting, Trondheim, Norway, October 2018
slides
Machine Learning & Ethics
International Journal of Cancer, Editor’s Meeting, Heidelberg, June 2017
slides
Introduction to Generative Adversarial Networks
karlsruhe.ai meetup, Karlsruhe, April 2017
slides
External Projects
The Post Binary
A 3-day conference on Artificial Intelligence in Art & Design held at the Museum Angewandte Kunst Frankfurt am Main
Co-Organizer, November 2018
website
heidelberg.ai
A Seminar Series on Machine Learning & Artificial Intelligence
Co-Initiator and -Organizer, now Program Committee Member, April 2017 - present
website youtube
Theses
Semantic Segmentation of Ambiguous Images
Department of Computer Science, Karlsruhe Institute of Technology (KIT).
Advised by Prof. Rainer Stiefelhagen & Klaus Maier-Hein.
PhD Thesis, December 2019
pdf
Dalitz Analysis of B−→D+π−π−
Institute for Experimental Nuclear Physics, Karlsruhe Institute of Technology (KIT). Advised by Prof. Michael Feindt.
Master Thesis, June 2016
pdf
Density Functional Theory of Tetrahedrites
Computational Condensed Matter Group, Oregon State University & Institute of Nanotechnology, Karlsruhe Institute of Technology (KIT). Advised by Prof. Guenter Schneider & Prof. Ferdinand Evers.
Bachelor Thesis, July 2014
pdf APS abstract
Reviewing
NeurIPS
Main Conference: 2020, 2019
Workshops: Machine Learning for Structural Biology 2020 (mlsb.io)
MICCAI
Main Conference: 2017
Workshops: UNSURE 2019 (unsuremiccai.github.io)
ICML
Main Conference: 2019
MIDL
Main Conference: 2019
© 2018-2020 Simon Kohl