MATHEMATICS AND COMPUTER SCIENCE

Lampert Group

Computer Vision and Machine Learning

Today’s computer programs are “idiots savant”: software that is extremely good at a certain task, such as playing chess, is completely useless for most other tasks like searching a database, and vice versa. The Lampert group works on methods for computers to break out of this limitation by sharing information between different tasks.

Modern computer software adapts to its users, e.g. voice recognition software learns to understand its speaker better over time, and email programs learn which of all incoming emails are spam and should therefore be suppressed. However, this learning process happens independently for each task that the computer is meant to solve. The Lampert group develops and analyzes algorithms that allow computers to learn new tasks while making use of the knowledge acquired from previous tasks. A particular application area is automatic image understanding, whereby the goal of the software is to analyze the contents of a natural image and automatically answer questions such as: What objects are visible in the image? Where are they located? How do they interact?

Group Leader


On this site:


Team


Current Projects

Life-long visual learning | Transfer learning | Image understanding with weak supervision | Structured prediction and learning


Publications

Zimin A. 2018. Learning from dependent data, IST Austria, 92p. View

Kolesnikov A. 2018. Weakly-Supervised Segmentation and Unsupervised Modeling of Natural Images, IST Austria, 113p. View

Darrell T, Lampert C, Sebe N, Wu Y, Yan Y. 2018. Guest editors’ introduction to the special section on learning with Shared information for computer vision and multimedia analysis. IEEE Transactions on Pattern Analysis and Machine Intelligence. 40(5), 1029–1031. View

Ringbauer H, Kolesnikov A, Field D, Barton NH. 2018. Estimating barriers to gene flow from distorted isolation-by-distance patterns. Genetics. 208(3), 1231–1245. View

Kuzborskij I, Lampert C. 2018. Data-dependent stability of stochastic gradient descent. Proceedings of the 35 th International Conference on Machine Learning. ICML: International Conference on Machine Learning vol. 80. 2815–2824. View

View All Publications

Career

since 2015 Professor, IST Austria
2010 — 2015 Assistant Professor, IST Austria
2007 – 2010 Senior Research Scientist, Max Planck Institute for Biological Cybernetics, Tübingen, Germany
2004 – 2007 Senior Researcher, German Research Center for Artificial Intelligence, Kaiserslautern, Germany
2003 PhD, University of Bonn, Germany


Selected Distinctions

since 2015 Associate Editor in Chief of the IEEE Transactions on Pattern Analysis and Machine Intelligence (T-PAMI)
2012 ERC Starting Grant
2008 Best Paper Award, IEEE Conference for Computer Vision and Pattern Recognition (CVPR)
2008 Best Student Paper Award, European Conference for Computer Vision (ECCV)
2008 Main Prize, German Society for Pattern Recognition (DAGM)


Additional Information

Open Lampert group website



Back to Top