Large-Scale Learning of SVM Classifiers. Autumn Pattern Recognition
and Computer Vision Colloquium. November 1, 2007. Prague.
[pdf]
Discriminative Learning of Max-Sum Classifiers. Talk at Friedrich
Miescher Laboratory of Max Planck Society
IDA internal seminar, April 19, 2007.
[pdf]
A Short Tutorial on Subgradient Methods. IDA retreat, May 16, 2007.
[pdf]
A Short Tutorial on Subgradient Methods. IDA internal seminar, April 19, 2007.
[pdf]
2006
Learning Max-Sum Classifiers From Inconsistent Examples. Princess project
talk. September 26, 2006.
[pdf]
Application of Structured Support Vector Machines Classifers.
PR course, Stockholm. August 15, 2006.
[pdf]
EM Algorithm-based Semi-supervise Learning. PR course,
Stockholm. August 14, 2006.
[pdf]
Tuning Parameters of Max-Sum Problem From Inconsistent Training
Examples. Zukin seminar, July 6, 2006.
[pdf]
Large-scale Quadratic Programming Solvers Based on Algorithms From
Computational Geometry. CMP internal seminar. March 21, 2006.
[pdf]
Large-scale Quadratic Programming Solvers Based on Algorithms From
Computational Geometery. One Day Optimization Colloquium, February 13, 2006. CTU
FEE Prague.
[pdf]
Quadratic Programming Solvers based on algorithms from
computational geometry. Job talk in IDA.FISRT. January 4, 2006. Berlin.
[pdf]
2005
Discriminative approach to learning Neyman-Pearson's strategy.
XXXVI Annnual Conference of the Italian Operational Research Society. September
7, 2005. Camerino, Italy.
[pdf]
Sequential Coordinate-wise Algorithm for the Non-negative Least Squares
Problem. Talk at the CAIP 2005 conference. September, 2005. Paris.
[pdf]
License plate character segmentation using Hidden Markov Chains.
Talk at the DAGM 2005 conference. September 2, 2005.
Wien. [pdf]
Sequential Quadratic Programming Solvers. Talk at the PRINCESS
meeting. Moscow. April 25, 2005. [pdf]
2004
Sequential Quadratic Programming Solvers. Talk at the Connex
meeting. Graz, Austria. 2004. [pdf]
Greedy Kernel Principal Component Analysis. Talk at the Dagstuhl seminar
on "Geometric Properties from Incomplete Data", Schloss Dagstuhl, Wadern,
Germany. 21.03.2004 - 26.03.2004. [pdf]
2003
Greedy Kernel Principal Component Analysis. Invited talk at the Max
Planck Institute for Biological Cybernetics, Department of Empirical
Inference for Machine Learning and Perception. Tuebingen, Germany.
December 17, 2003.
[pdf]
Introduction to pattern recognition. Course on PR at the FEE CTU
Prague. November 2003.
[ppt]
Greedy Algorithm for a Training Set Reduction in the Kernel Methods.
Computer Analysis of Images and Patterns. Groningen, The Netherlands,
August 25-27, 2003.
[pdf]
Notes to the Neal and Hinton's paper: A View of the EM Algorithm that
Justifies Incremental, Sparse, and other Variants.
Reading group on the EM algorithm which took place in CMP, spring 2003.
[ps]
Training Set Approximation for Kernel Methods
Computer Vision Winter Workshop. Valtice, Czech Republic. February 6, 2003.
[pdf]
2002
Multi-class Support Vector Machines.
Invited talk at the ETH Zurich. December 12, 2002.
[pdf]
Generalization of the Schlesinger-Kozinec's algorithm for Support Vector Machines.
Invited talk at the Dresden TU. May 6, 2002.
[pdf]
2001
A contribution to the Schlesinger's algorithm separating mixtures of Gaussians .
Conference: Computer Analysis of Images and Patterns. Warsaw, Poland.
September 5, 2001.
[pdf]
A Simple Learning Algorithm for Maximal Margin Classifier . Workshop on
the Kernel and Subspace Methods for Computer Vision adjoint to ICANN.
Austria. August 25, 2001.
[pdf]
New features of the Statistical Pattern Recognition Toolbox. Conference:
Computer Vision, Computer Graphics and Photogrammetry -- a Common Viewpoint,
OeAGM. Austria. June 8, 2001.
[pdf]
Convergence of the EM algorithm to the global maximum.
Introduction talk at the PRIP, Austria. March 29, 2001.
[pdf]