Machine Learning and Scientific Data Analysis

Contact

Head of group:

Dr. Niels Landwehr
Email: landwehr@cs.uni-potsdam.de
Phone: +49 331 977 3000
Fax: +49 331 977 3022

Secretary:

Alexandra Roy, roy@cs.uni-potsdam.de.

Postal Address:

Machine Learning and Scientific Data Analysis
University of Potsdam
Department of Computer Science
August-Bebel-Str. 89
14482 Potsdam, Germany
How to find us on Google Maps.

Overview

The research group Machine Learning and Scientific Data Analysis is funded by the German Research Foundation (DFG) under the Emmy Noether-Programme and headed by Niels Landwehr. The group is affiliated with the Machine Learning Research Group headed by Tobias Scheffer.

The research focus of the group is in machine learning, algorithmic data analysis, and statistical model building from observational data in the sciences. We specifically study data that have complex distributional characteristics, such as data that vary spatially or temporally. In collaboration with researchers from seismology, cognitive psychology, and remote sensing, we study practical data analysis and model building problems.

News

• The paper Varying-Coefficient Models for Geospatial Transfer Learning was accepted at the Machine Learning Journal (April 2017).
• The paper A Semiparametric Model for Bayesian Reader Identification was accepted at EMNLP-2016 (July 2016).
• The paper A Non-Ergodic Ground Motion Model for California with Spatially Varying Coefficients was accepted at the Bulletin of the Seismological Society of America (July 2016).
• The paper Learning to Identify Concise Regular Expressions that Describe Email Campaigns was accepted at the Journal of Machine Learning Research (February 2015).
• Niels Landwehr will serve as Program Co-Chair of the 2016 European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases in Riva del Garda, Italy (September 2014).
• The paper A Model of Individual Differences in Gaze Control During Reading was accepted at EMNLP-2014 (July 2014).
• The paper Joint Prediction of Topics in a URL Hierarchy was accepted at ECML-2014 (June 2014).
• Invited presentation at IJCAI-2013, Best Papers in Sister Conferences Track, for our paper Active Evaluation of Ranking Functions based on Graded Relevance (August 2013).
• The paper Logistic Model Trees receives the ECML-2013 Test of Time Award (August 2013).
• The paper Active Evaluation of Ranking Functions based on Graded Relevance was accepted at the Machine Learning Journal (April 2013).