Publications of Ulrike von Luxburg

The papers are roughly ordered by topic and year of submission. Papers that did not go though a serious peer-review process are cited in grey font. Bibtex entries can be found here.


M. Alamgir and U. von Luxburg: Multi-agent random walks for local clustering. International Conference on Data Minig (ICDM), 2010. pdf

U. von Luxburg and A. Radl and M. Hein: Getting lost in space: Large sample analysis of the commute distance. Neural Information Processing Systems (NIPS), 2010. pdf (paper with supplement)     An annoying typo

U. von Luxburg. Clustering stability: an overview. Foundations and Trends in Machine Learning 2 (3), 235-274, 2010. pdf

I. Guyon, U. von Luxburg, R. Williamson: Clustering: Science or Art? Opinion paper for the NIPS workshop "Clustering: Science or Art", 2009. pdf

U. von Luxburg. Evidenzkriterien in der Informatik (in German). In: E. Engelen and C. Fleischhack and G. Galizia and K. Landfester (Eds): Heureka: Evidenzkriterien in den Wissenschaften. Springer, Berlin, 2010. pdf
This article has been written for a book that compares how different scientific disciplines establish their results and what kind of evidence one has to give before a result is considered to be true. link to the book

U. von Luxburg and B. Schölkopf. Statistical Learning Theory: Models, Concepts, and Results. In: D. Gabbay, S. Hartmann and J. Woods (Eds). Handbook of the History of Logic, vol 10: Inductive Logic. In Press. pdf
This is a high level overview paper on statistical learning theory. We try to convey the major ideas and insights of SLT, rather than to dive into technical details. It is intended for a broader audience - we do not assume that the reader is particularly familiar with machine learning.

S. Bubeck, M. Meila, U. von Luxburg. How the initialization affects the stability of the k-means algorithm. Draft, 2009. link

S. Bubeck and U. von Luxburg. Nearest Neighbor Clustering: A Baseline Method for Consistent Clustering with Arbitrary Objective Functions. JMLR 10, 657-698, 2009. pdf
(Note: a previous version of this manuscript was called "Overfitting of clustering and how to avoid it").

M. Maier, U. von Luxburg, M. Hein: Influence of graph construction on graph-based clustering measures. In: D. Koller and D. Schuurmans and Y. Bengio and L. Bottou (Eds.): Advances in Neural Information Processing Systems (NIPS) 22, 2009. paper (pdf) and supplement (pdf)
(For this paper Markus Maier received the NIPS best student paper award. ).

U. von Luxburg, S. Bubeck, S. Jegelka, M. Kaufmann: Consistent Minimization of Clustering Objective Functions. In: J. Platt and D. Koller and Y. Singer and S. Roweis (editors): Advances in Neural Information Processing Systems (NIPS) 21, MIT Press, Cambridge, MA. 2008
paper (pdf), and supplement (pdf).

S. Ben-David and U. von Luxburg: Relating clustering stability to properties of cluster boundaries. In: R. Servedio and T. Zhang (Eds.): Proceedings of the 21st Annual Conference on Learning Theory (COLT), pp. 379-390. Springer, Berlin, 2008. pdf

M. Maier and M. Hein and U. von Luxburg. Optimal construction of k-nearest neighbor graphs for identifying noisy clusters. Theoretical Computer Science 410, p. 1749-1764, 2009. preprint as pdf

M. Maier, M. Hein, U. von Luxburg. Cluster identification in nearest neighbor graphs. In: Marcus Hutter and Rocco A. Servedio and Eiji Takimoto (Eds): Algorithmic Learning Theory (ALT) 18, pp. 196--210. Springer, 2007. pdf (conference paper), pdf (technical report with proof details)
(For this paper Markus Maier received the ALT best student paper award. ).

U. von Luxburg and V. Franz. A Geometric Approach to Confidence Sets for Ratios: Fieller's Theorem, Generalizations, and Bootstrap. Statistica Sinica 19 (3), pp. 1095 - 1117, 2009
preprint of the paper (pdf) and supplement (pdf)
[A preliminary version of this paper with less results appeared as: Confidence sets for ratios: A purely geometric approach to Fieller's theorem. Technical Report 133, Max Planck Institute for Biological Cybernetics, 2004. pdf]

U. von Luxburg. A Tutorial on Spectral Clustering. Statistics and Computing 17(4): 395-416, 2007. paper (pdf) and some typos (txt).
[A previous version of this paper appeared as Technical Report 149, Max Planck Institute for Biological Cybernetics, 2006. ] There also exists a video lecture where I give a tutorial on spectral clustering (and other clustering algorithms). And here is a nice matlab demo which can be used to play with spectral clustering (written by Matthias Hein and me).

S. Ben-David, U.von Luxburg, D. Pal: A Sober Look on Clustering Stability. In: G. Lugosi and H. Simon, editors, Proceedings of the 19th Annual Conference on Learning Theory (COLT), pages 5 - 19, Springer, 2006. pdf
(For this paper David Pal received a COLT best student paper award).

Matthias Hein, Jean-Yves Audibert, Ulrike von Luxburg. Graph Laplacians and their Convergence on Random Neighborhood Graphs. JMLR 8:1325--1370, 2007. pdf,

M. Hein, J.-Y. Audibert, and U. von Luxburg. From Graphs to Manifolds - Weak and Strong Pointwise Consistency of Graph Laplacians. In: P. Auer and Ron Meir, editors, Proceedings of the 18th Annual Conferecnce on Learning Theory (COLT), pages 470-485. Springer, 2005. pdf
(For this paper Matthias Hein received a COLT best student paper award).

U. von Luxburg, S. Ben-David. Towards a statistical theory for clustering. Presented at the PASCAL Workshop on Statistics and Optimization of Clustering Workshop 4-5 July 2005, London, U.K. pdf.
[This opinion paper suggested some kind of "research program". Note that in the meantime, some of the questions have already been solved, while some of the other questions turned out to be not so useful after all...]

U. von Luxburg, M. Belkin, and O. Bousquet. Consistency of spectral clustering. Annals of Statistics, 36 (2), 555-586, 2008 pdf
[Previous version appeared as Technical Report 134, Max Planck Institute for Biological Cybernetics. ]

U. von Luxburg, O. Bousquet, and M. Belkin. Limits of spectral clustering. In Lawrence K. Saul, Yair Weiss, and Leon Bottou, editors, Advances in Neural Information Processing Systems (NIPS) 17. MIT Press, Cambridge, MA, 2005. pdf
(For this paper I received the NIPS outstanding student paper award).

U. von Luxburg, O. Bousquet, and M. Belkin. On the convergence of spectral clustering on random samples: the normalized case. In J. Shawe-Taylor and Y. Singer, editors, Proceedings of the 17th Annual Conference on Learning Theory (COLT), pages 457-471. Springer, 2004. pdf

U. von Luxburg and O. Bousquet. Distance-based classification with Lipschitz functions. Journal for Machine Learning Research, 5:669-695, 2004. pdf

U.von Luxburg and O.Bousquet. Distance-based classification with Lipschitz functions. In B.Schölkopf and M.K. Warmuth, editors, Proceedings of the 16th Annual Conference on Learning Theory (COLT), pages 314-328. Springer, 2003. pdf
(For this paper I received the COLT Mark Fulk best student paper award).

U. von Luxburg, O.Bousquet, and B.Schölkopf. A compression approach to support vector model selection. Journal for Machine Learning Research, 5:293-323, 2004. pdf
[An earlier version of this paper appeared as Technical Report 101, Max Planck Institute for Biological Cybernetics].


Edited Books

O. Bousquet, U. von Luxburg, and G.Rätsch, editors. Advanced Lectures on Machine Learning, volume 3176 of Springer Lecture Notes in Artificial Intelligence. Springer Verlag, Heidelberg, 2004. link


PhD Thesis

U. von Luxburg. Statistical Learning with Similarity and Dissimilarity Functions. PhD thesis, Technical University of Berlin, 2004. pdf