Principal Investigator Bastian Rieck
AIDOS lab 

Our world is full of phenomena that occur at multiple scales. Biomedical research, for instance, commonly observes complex systems at different resolutions, ranging from the macroscopic to the microscopic. Zooming in provides us with the 'fine print' (e.g. individual neurons in a brain), while zooming out lets us see the 'big picture' (e.g. locally-connected networks of neurons, or areas in the brain). For many applications, there is not just one specific scale to consider—relevant features might occur on multiple scales and a priori information about their suitability for a specific task is typically lacking.

With noise being an inevitable part of such investigations, we need tools that enable robust multi-scale analyses. Our research agenda is to create, cultivate, and critique such tools based on topological machine learning techniques, with a specific focus on healthcare topics.

Our Aims

Develop novel machine learning algorithms based on topological concepts that are (I) aware of the multi-scale nature of complex biomedical data sets, (II) robust to noise, and (III) interpretable.

Dr. Bastian Rieck

Principal Investigator, aidos lab 

Bastian received both his M.Sc. degree in mathematics (graduated with distinction) and his Ph.D. ('summa cum laude') in computer science from Heidelberg University in Germany. He subsequently worked as a (senior) postdoctoral researcher at the Machine Learning and Computational Biology Lab of ETH Zurich in Switzerland. Bastian's research primarily focuses on the development of topological machine learning techniques in the context of healthcare applications. He also has a keen interest in developing techniques that improve our understanding of neural networks.
 

YouTube Channel

Factsheet

Dr. Bastian Rieck 

Positions and Career

2021 - present 
Principal Investigator at Helmholtz Pioneer Campus, Helmholtz Zentrum München, Germany

2020- 2021
Senior assistant at ETH Zurich, Switzerland

2018-2019
Postdoctoral researcher at ETH Zurich, Switzerland

2017
Researcher at Heidelberg University, Germany

2015-2017
Research assistant at Kaiserslautern University, Germany

2011-2015
Research assistant at Heidelberg University, Germany.

 

 

2011-2017
Ph.D. in Computer Science at Heidelberg University, Germany

2005-2011
M.Sc.1 in Mathematics at Heidelberg University, Germany

2021
Outstanding reviewer (top 10%) for ICML
Outstanding reviewer (among the top 500 reviewers) for ICLR

2020
Top reviewer for ICML 2020

2019
Outstanding reviewer for the ECML PKDD5 2019 Journal Track
Outstanding reviewer (among the top 400 reviewers) for NeurIPS
Outstanding reviewer (top 5%) for ICML 

2018
Outstanding reviewer (among the top 200 reviewers) for NeurIPS 
Outstanding reviewer (among the top 100 reviewers) for ICML 

2017
Award for the best extended abstract at TopoInVis

2011-2014
Research scholarship, Heidelberg Graduate School of Mathematical and Computational Methods for the Sciences (HGS MathComp), Heidelberg University, Germany

 

 

Selected Publications

Back to the basics with inclusion of clinical domain knowledge — A simple, scalable, and effective model of Alzheimer’s Disease classification.

S. C. Brüningk†, F. Hensel†, L. Lukas, M. Kuijs, C. R. Jutzeler‡, and B. Rieck
Proceedings of the 6th Machine Learning for Healthcare Conference. Proceedings of Machine Learning Research, 2021. In press.


Filtration Curves for Graph Representation

O’Bray†, B. Rieck†, and K. Borgwardt
Proceedings of the 27th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining (KDD), 2021. doi:10.1145/3447548.3467442. In press.

More Details

Early Prediction of Sepsis in the ICU Using Machine Learning: A Systematic Review

M. Moor†, B. Rieck†, M. Horn, C. R. Jutzeler‡, and K. Borgwardt‡
in Medicine 8, 2021. doi: 10.3389/fmed.2021.607952.

More Details

A Survey of Topological Machine Learning Methods

F. Hensel, M. Moor, and B. Rieck
Frontiers in Artificial Intelligence 4, 2021. doi: 10.3389/frai.2021.681108.

More Details

Stable Topological Signatures for Metric Trees through Graph Approximations

R. Vandaele, B. Rieck, Y. Saeys, and T. De Bie
Pattern Recognition Letters 147, pp. 85–92, 2021. doi: 10.1016/j.patrec.2021.03.035

More Details

Uncovering the Topology of Time-Varying fMRI Data using Cubical Persistence

B. Rieck†, T. Yates†, C. Bock, K. Borgwardt, G. Wolf, N. Turk-Browne‡, and S. Krishnaswamy‡
Advances in Neural Information Processing Systems (NeurIPS), Volume 33, pp. 6900–6912, 2020. Accepted as a spotlight presentation at NeurIPS (top 3% of all submissions). arXiv: 2006.07882 [q-bio.NC]

More Details

Graph Kernels: State-of-the-Art and Future Challenges

K. Borgwardt, E. Ghisu, F. Llinares-López, L. O’Bray, and B. Rieck
Foundations and Trends® in Machine Learning 13:5–6, pp. 531–712, 2020. doi: 10.1561/2200000076. arXiv: 2011.03854 [cs.LG]
 

More Details

Contact Us

HPC contact

Contact us
I consent to the information and contact details provided by myself being used by Helmholtz Zentrum München , in order to contact me for communication purposes and address my query. This is especially applicable for the use of my e-mail address and potentially my phone number. I know that I can revoke my consent to the collection, use and storage of my personal data at any time by sending my revocation to presse(at)helmholtz-muenchen.de.
For any queries regarding the use of my personal data on this website, please see the data protection statement.
For any further enquiries regarding your personal data, please contact our Data Protection Officer at datenschutz(at)helmholtz-muenchen.de

Helmholtz Pioneer Campus
Helmholtz Zentrum München
Deutsches Forschungszentrum für Gesundheit und Umwelt (GmbH)

Ingolstädter Landstr. 1
85764 Neuherberg
Germany