Invited Speakers


We proudly announce these three invited speakers:


Speaker Time Title
Levent Tunçel Monday 15th
9:30-10:30
Semidefinite Programming and Lift-and-Project Methods in Combinatorial Optimization
Thomas Breitling Tuesday 16th
9:00-10:00
The Challenges of Multidisciplinary Optimization in an Automotive Development Process
Arndt von Haeseler Wednesday 17th
9:00-10:00
Phylogenetic Tree Reconstruction:
A Challenging Optimization Problem from Nature.






Semidefinite Programming and Lift-and-Project Methods in Combinatorial Optimization

Levent Tunçel

Department of Combinatorics & Optimization
Faculty of Mathematics
University of Waterloo, Canada

Monday 15th, 9:30-10:30

Abstract

Semidefinite programming (SDP) is one of the most active research areas in optimization where the variables are represented by symmetric matrices some of which are required to be positive semidefinite. This framework ends up being a powerful generalization of linear programming problems. Lift-and-project methods provide universal ways of applying SDP techniques to combinatorial optimization problems. I will go over the theory of these methods first. Then, I will turn to their performance on various combinatorial optimization problems with a special focus on the maximum weight stable set problem. I will conclude the talk with a very brief discussion of generalizations of these methods to deal with optimization of polynomials over semi-algebraic sets.

Biosketch

Levent Tunçel is currently a Professor in the Department of Combinatorics and Optimization, Faculty of Mathematics, University of Waterloo, Canada. He received a B.Sc. in Industrial Engineering from Dokuz Eylül University and a M.Sc. in Industrial Engineering from the Middle East Technical University in Turkey. Then he received a M.S. and a Ph.D. in Operations Research from Cornell University in the USA. He served as an associate editor or guest editor for Operations Research, Mathematical Programming and Transactions on Operational Research. Currently, he is serving on the editorial board of SIAM Journal on Optimization. Tunçel's research program is in mathematical optimization and mathematics of operations research, it aims at understanding the structure of optimization problems and designing provably efficient methods for solving them.


back to top



The Challenges of Multidisciplinary Optimization in an Automotive Development Process

Thomas Breitling

Daimler AG
Sindelfingen, Germany

Tuesday 16th, 9:00-10:00

Abstract

Many methods in the scope of the PPSN conference are state of the art and are already available in the optimization kernels of commercial tools. The evolutionary principles adapted by these methods, however, were also applied to the algorithms themselves: In the beginnings of Evolutionary Computation, many variants emerged and competed with already established ones, but only a few prevailed and are considered as mainstream nowadays. Now the selection pressure starts rising again: With the increasing complexity of products and product development processes, there is a demand for methods which allow the efficient and reliable validation of product designs regarding a great number of possibly conflicting requirements. The automation of these validation processes exhibits new challenges for optimization tools which are summarized under the term of Multidisciplinary Optimization. By the example of the Daimler passenger car development, the talks describes the special requirements optimization methods have to fulfill in the near future, especially the limited number of single design verifications due to the long simulation times of certain disciplines.

Biosketch

Thomas Breitling received a degree in Chemistry and Biology at the Eberhard-Karls-University of Tübingen, Germany. Then he received his diploma and PhD in General Mechanical Engineering at the University of Karlsruhe, Germany. He worked in different departments for Daimler AG, Mercedes-Benz Cars in Sindelfingen.
At the moment he is the head of Active Safety, Vehicle Dynamics and Energy Management at Mercedes-Benz Cars Development at Daimler AG in Sindelfingen, Germany. In 2002 he became Honorary Professor at Karlsruhe University, Germany.


back to top



Phylogenetic Tree Reconstruction:
A Challenging Optimization Problem from Nature.

Arndt von Haeseler

Center for Integrative Bioinformatics Vienna (CIBIV)
Max F. Perutz Laboratories (MFPL)
Vienna, Austria

Wednesday 17th, 9:00-10:00

Abstract

We will give a brief overview about the state of the art of phylogenetic tree reconstruction and will outline the problems biologists and computer scientists face, when trying to reconstruct phylogenetic trees from the vast amount of molecular biology data available today. Finally, we will point out some recent developments in phylogeny reconstruction and explain some open problems.

Biosketch

Arndt von Haeseler was appointed scientific director of the Center for Integrative Bioinformatics Vienna (CIBIV) in September 2005 and professor for bioinformatics. He studied biology and mathematics and received his PhD in Mathematics at the University of Bielefeld. Then he did his post-doc at the University of Southern California (Los Angeles) with Mike Waterman. Subsequently he was an assistant professor at the University of Munich, an associate professor at the Max Planck Institute for evolutionary Anthropology, and full professor for Bioinformatics at the University of Düsseldorf. His research interests are: phylogenetic tree reconstruction, modelling evolution, and population genetics, algorithms for bioinformatics, biodiversity parallel computation.

back to top