Volker Roth

 Tel: +41 1 63 23179
 E-Mail:

Short Bio:
I received a Diploma in Physics from the University of Bonn in 1997.  After that, I changed to the Institute of Computer Science and received a Ph.D. in Computer Science in 2001. From 2001 till fall 2003 I was a postdoc researcher at the Computer Vision an Pattern Recognition Group at the institute of Computer Science III in Bonn.
Currently I am with the Machine Learning Group at ETH Zurich, headed by Prof. J. Buhmann. In October 2005 I received the "Venia Legendi" (Habilitation as "Privatdozent") for the field "computational science" from ETH Zurich.

In October 2007 I will be joining University of Basel as Assistant Professor for Biomedical Data Analysis.

Research Areas:
Support Vector Machines and kernel-based algorithms. Unsupervised learning and clustering. Data fusion. Applications of Machine Learning in the area of Computational Life Sciences.

Software

• GenLASSO: the Generalized LASSO Classifier.
• kLASSO: the kernelized LASSO for least-squares regression. The tar file includes a short description.
• OC-KFD: Outlier Detection with One-Class Kernel Fisher Discriminants.

List of Publications

• Time-Series Alignment by Non-Negative Multiple Generalized Canonical Correlation Analysis. Bernd Fischer , Volker Roth and Joachim M. Buhmann. BMC Bioinformatics, to appear.
• The kernelHMM: Learning Kernel Combinations in Structured Output Domains. Volker Roth, Bernd Fischer. Pattern Recognition--DAGM 2007, to appear.
• NONNEGATIVE CCA FOR AUDIOVISUAL SOURCE SEPARATION. Christian Sigg, Bernd Fischer, Bjorn Ommer, Volker Roth and Joachim M. Buhmann. MACHINE LEARNING FOR SIGNAL PROCESSING--MLSP 2007, to appear.
• Time-Series Alignment by Non-Negative Generalized Canonical Correlation Analysis. B. Fischer, V. Roth, J.M. Buhmann. Computational Intelligence Methods for Bioinformatics and Biostatistics--CIBB 2007, to appear.
• Improved Functional Prediction of Proteins by Learning Kernel Combinations in Multilabel Settings. Volker Roth, Bernd Fischer. BMC Bioinformatics, Vol. 8 Suppl.2:S12, 2007.
• Semi-Supervised LC/MS Alignment for Differential Proteomics. Bernd Fischer, Jonas Grossmann, Volker Roth, Wilhelm Gruissem, Sacha Baginsky, Joachim M. Buhmann. ISMB 2006. Bioinformatics 22(14):e132-e140.
• The Science of "Fingerprinting" Bees. V. Steinhage, S. Schroeder, V. Roth, A.B. Cremers, W. Drescher, D. Wittmann. German Research 28(1):19-21, Wiley-VCH, 2006.
• Improved Functional Prediction of Proteins by Learning Kernel Combinations in Multilabel Settings. Volker Roth, Bernd Fischer. In: Probabilistic Modeling and Machine Learning in Structural and Systems Biology (PMSB 2006), University of Helsinki, Technical Report B-2006-4, 2006.
• Exploiting Low-level Image Segmentation for Object Recognition. Volker Roth, Bjorn Ommer. In: Pattern Recognition--DAGM'06. Springer, volume 4174 of LNCS, pages 11--20, 2006.
• On the Information and Representation of Non-Euclidean Pairwise Data. Julian Laub, Volker Roth, Joachim M. Buhmann, Klaus-Robert Mueller. Pattern Recognition 39(10):1815-1826, 2006.
• NovoHMM: A Hidden Markov Model for de Novo Peptide Sequencing. Bernd Fischer, Volker Roth, Franz Roos, Jonas Grossmann, Sacha Baginsky, Peter Widmayer, Wilhelm Gruissem, and Joachim M. Buhmann. Analytical Chemistry, 77(22):7265 - 7273, 2005.
• Kernel Fisher Discriminants for Outlier Detection. Volker Roth. Neural Computation, 18(4), 2006.
• A Hidden Markov Model for de Novo Peptide Sequencing. Bernd Fischer, Volker Roth, Joachim M. Buhmann. NIPS*17, pages 457-464. MIT Press, Cambridge, MA, 2005.
• Outlier Detection with One-class Kernel Fisher Discriminants. Volker Roth. NIPS*17, pages 1169-1176. MIT Press, Cambridge, MA, 2005.
• Feature Selection in Clustering Problems. Volker Roth, Tilman Lange. NIPS*16, 2004.
• Adaptive Feature Selection in Image Segmentation. Volker Roth, Tilman Lange. Pattern Recognition--DAGM'04, Springer, volume 3175 of LNCS, 2004.
• Clustering with the Connectivity Kernel. Bernd Fischer, Volker Roth, Joachim M. Buhmann. NIPS*16, 2004.
• Stability-Based Validation of Clustering Solutions. Tilman Lange, Volker Roth, Mikio L. Braun and Joachim M. Buhmann,   Neural Computation, 16(6):1299 -- 1323, 2004.
• Bayesian Class Discovery in Microarray Datasets. Volker Roth and Tilman Lange,   IEEE Transactions on Biomedical Engineering, VOL. 51, NO.5, May 2004.
• Optimal Cluster Preserving Embedding of Non-Metric Proximity Data. Volker Roth, Julian Laub, Motoaki Kawanabe, Joachim M. Buhmann, IEEE Transactions on Pattern Analysis and Machine Intelligence, Vol. 25, NO.12, December 2003.
• The Generalized LASSO. Volker Roth,   IEEE Transactions on Neural Networks, Vol. 15, NO. 1, January 2004.
• Going metric: Denoising pairwise data NIPS*15, 2003. Volker Roth, Julian Laub, Joachim M. Buhmann, Klaus-Robert Mueller.
• Stability-Based Model Selection. NIPS*15, 2003. Tilman Lange, Mikio Braun, Volker Roth, Joachim Buhmann.
• The Generalized LASSO: a wrapper approach to gene selection for microarray data. Volker Roth, University of Bonn, Dep. Computer Science III, Tech. Report IAI-TR-2002-8., 2002
• Stability-Based Model Order Selection in Clustering with Applications to Gene Expression Data.  Volker Roth, Mikio Braun, Tilman Lange, Joachim M. Buhmann. J.R. Dorronsoro (Ed.): Artificial Neural Networks - ICANN 2002, Springer, LNCS 2415.
• A Resampling Approach to Cluster Validation. Volker Roth, Tilman Lange, Mikio Braun, Joachim M. Buhmann. Computational Statistics--COMPSTAT 2002.
• The New Key to Bees: Automated Identification by Image Analysis of Wings. S. Schröder, D. Wittmann, W. Drescher, V. Roth, V. Steinhage and A.B. Cremers. In: Pollinating Bees: The Conservation Link Between Agriculture and Nature, Proc. Workshop on Conservation and Sustainable Use of Pollinators in Agriculture, S. Paulo, Brazil, p. 209-218, 2002.
• Pairwise Coupling for Machine Recognition of Hand-Printed Japanese Characters, Volker Roth and Koji Tsuda, Computer Vision and Pattern Recognition--CVPR 01, p. 1120--1125.
• Sparse Kernel Regressors. V. Roth, In: Dorfner, Bischof, Hornik (eds.),  Artificial Neural Networks--ICANN 2001,  p. 339-346 , Springer, LNCS 2130.
• Ph.D. thesis, Kernel Methods for Regression and Classification, Fortschrittsberichte VDI,  Reihe 10, Nr. 671, VDI Verlag Duesseldorf
• Probabilistic Discriminative Kernel Classifiers for Multi-class Problems. V. Roth , In: Radig B., Florczyk S. (eds.), Pattern Recognition--DAGM'01, p. 246-253, Springer, LNCS 2191
• Automatisierte Artenbestimmung von Insekten durch Bildanalyse ,   V. Roth, A. Pogoda, V. Steinhage, S. Schröder, Kuenstliche Intelligenz, 2000(1), p. 48-49, arenDTaP Verlag, Bremen, 2000.
• Nonlinear Discriminant Analysis Using Kernel Functions. Roth V. & Steinhage V.,  In: S.A. Solla and T.K. Leen and  K.-R. Mueller (eds.) , Advances in Neural Information Processing Systems--NIPS*12, p. 568-574,1999 .     A longer version has been published as Technical Report Nr. IAI-TR-99-7, University of Bonn, Informatik III, ISSN 0944-8535
• Pattern Recognition Combining Feature- and Pixel-based Classification Within a Real World Application. Roth V., Pogoda A., Steinhage V., Schröder S., 1999 In: Mustererkennung 1999--DAGM'99, Bonn, Informatik aktuell, Springer
• Pattern Recognition Combining De-Noising and Linear Discriminant Analysis within a Real World Application. Roth V., Steinhage V., Schröder S., Cremers A.B., Wittmann D., 1999, Computer Analysis of Images and Patterns, CAIP 99, Ljubljana, Lecture Notes in Computer Science 1689, Springer
• Computergestützte Klassifikation von Wildbienen mit Methoden der Bildanalyse . Roth V., Schröder S., Cremers A.B., Drescher W., Steinhage V., Wittmann D. (in german), KI/UI-98: workshop Intelligente wissensbasierte Systeme in der Umwelttechnik, Forschungszentrum Karlsruhe, Wissenschaftliche Berichte FZKA 6252, ISSN 0947-8620.