Ulrike von Luxburg

from Wikipedia, the free encyclopedia

Ulrike von Luxburg (* 1975 in Regensburg ) is a German computer scientist and head of the Theory of Machine Learning working group at the Wilhelm Schickard Institute of the Faculty of Mathematics and Natural Sciences at the Eberhard Karls University in Tübingen .

Career

Von Luxburg studied mathematics at the University of Konstanz , the University of Grenoble and the University of Tübingen, where she received her Dipl.-Math. completed. From 2002 to 2004 she worked as a doctoral candidate at the Department of Empirical Inference at the Max Planck Institute for Biological Cybernetics in Tübingen . In 2004 she was at Stefan Jähnichen at the Technical University of Berlin with a thesis on "Statistical Learning with Similarity and dissimilarity functions" to Dr. rer. nat. PhD.

From 2005 to 2006, von Luxburg headed the Data Mining research group at the Fraunhofer Institute for Integrated Publication and Information Systems (IPSI) in Darmstadt . From 2007 to 2012, von Luxberg headed the learning theory research group at the Max Planck Institute for Intelligent Systems in Tübingen. In 2012 she followed the call to a Heisenberg Professorship for Machine Learning at the Faculty of computer science of the University of Hamburg . 2015 she was appointed as professor of theoretical computer science and learning theory at the WSI Institute of the University of Tübingen, where she has since the Working Group Theory of Machine Learning guides.

With her research activities on the subject of machine learning, von Luxburg and her work group develop algorithms that enable a fundamental analysis of large amounts of complex data in order to find connections or answer specific questions. In this way, enormous amounts of data, which are recorded, collected and stored in the course of numerous applications in science and industry, can be used and evaluated.

One of the strengths of machine learning is the ability to model complex structures within the data. Graphs can be used here to model relationships and structures: the interaction of proteins in a metabolic network, the structure of complex chemical bonds, the interaction of neurons in the brain, dependencies between linked pages on the Internet as well as social interactions between people or communication between Sensors in a network.

The focus of her research is to combine methods from statistics and computer science in order to theoretically evaluate algorithms for machine learning on graphs. The core of the investigations is how algorithms react to certain data sets. In machine learning, the question is to what extent a result obtained from a certain data set can be regarded as typical for the underlying population, i.e. H. whether it is significant or just a random artifact with no further meaning.

Awards

Publications (selection)

Web links

Individual evidence

  1. ↑ Ulrike von Luxburg: Statistical Learning with Similarity and Dissimilarity Functions . ( Dissertation ) 2004
  2. ^ Profile of Ulrike vo Luxburg at the Young Academy
  3. ^ Ulrike von Luxburg in the database of renowned academics AcademiaNet
personal data
SURNAME Luxburg, Ulrike von
BRIEF DESCRIPTION German computer scientist and university professor
DATE OF BIRTH 1975
PLACE OF BIRTH regensburg