Logo Uni Bremen

Center for Industrial Mathematics

ZeTeM > Working Groups > WG Discrete Optimization

Contact Sitemap Impressum [ English | Deutsch ]

WG Discrete Optimization

Administration: Prof. Dr. Daniel Schmand

The Discrete Optimization group develops efficient algorithms for application-motivated mathematical optimization problems. A special focus of the group is the computation of solutions with additional properties. In Algorithmic Game Theory, stable equilibrium states of complex systems with strategically acting agents are studied. In robust optimization, solutions are sought whose quality is robust to a changed initial situation. Typical applications are in traffic planning and optimization, logistics, roster optimization, automative systems, and artificial intelligence. We work closely in interdisciplinary collaborations with other scientists and industrial companies.
Research Areas
Algorithmic Game Theory

Many every-day processes can seen as a game between autonomous interacting players, where each player acts stategically in order to pursue her own objectives. Use-cases are distributed systems, auctions, online-markets, resource allocation, traffic routing, and sports. One goal of Algorithmic Game Theory is to analyze the efficiency and existence of equilibria in such games.

Discrete Optimization

In the research area Discrete Optimization, methods, algorithms and software are developed to handle discrete optimization problems. Application areas include the planning of network and transportation systems. Many application examples can be modeled as graphs. An easy application example from this area is the construction of the least cost connected network; i.e., the construction of a minimum spanning tree given a graph. Kruskal's algorithm is an algorithm that generates such a solution:

Online Algorithms

An online algorithm is one that can process its input piece-by-piece in a serial fashion, i.e., in the order that the input is fed to the algorithm, without having the entire input available from the start. In contrast, an offline algorithm is given the whole problem data from the beginning and is required to output an answer which solves the problem at hand.
In particular, our research group studies variants of the secretarial problem, where the task is to select the best candidate from a set of candidates considered individually and sequentially.

Latest news
FRICO 23, Eindhoven, Best talk award

Nicole Schröder and Torben Schürenberg attended FRICO 23 in Eindhoven from Aug. 14 - Aug. 18 and each gave a talk. FRICO (Future Research in Combinatorial Optimization) took place for the 26th time this year. 38 PhD students presented topics in discrete mathematics for 30 minutes each. The best talk award at FRICO 2023 went to Torben Schürenberg from University of Bremen for his talk "Hunting an invisible rabbit on an infinite graph in finite time".

Welcome Nicole Schröder and Vijayaragunathan Ramamoorthi!

Vijayaragunathan Ramamoorthi joined our working group as well as the groups of Nicole Megow and Sebastian Siebertz on Nov. 2. He will advance projects in the context of Minds-Media-Machines and connect the working groups. Welcome!

Welcome Nicole Schröder und Vijayaragunathan Ramamoorthi

Nicole Schröder joined our working group on Dec. 1. Welcome!

Vijayaragunathan Ramamoorthi joined our working group as well as the groups of Nicole Megow and Sebastian Siebertz on Nov. 2. He will advance projects in the context of Minds-Media-Machines and connect the working groups. Welcome!

New preprint @arXiv and new aper @ WINE 2022

Our preprint ''Prophet-Inequalities over Time'' (by Andreas Abels, Elias Pitschmann and Daniel Schmand) is now on arXiv.

Our paper ''Bicriteria Nash Flows over Time'' (by Tim Oosterwijk, Daniel Schmand, and Marc Schröder) is accepted at WINE 2022 (The 18th Conference on Web and Internet Economics).

Activities
Hosting of the Days on Computational Game Theory 2022 in Bremen, Germany.

The working group hosted the Days on Computational Game Theory at the University of Bremen on September 29 and 30, 2022. The workshop brought together researchers from different areas of algorithmic game theory in a friendly atmosphere to promote scientific exchange.

Lecture and exhibit at CAMPUS CITY 2022

From October 10 - 23, 2022, the University of Bremen presented itself in the city. Lectures, guided tours, exhibitions, workshops and much more could be visited. We were also present and brought our research area closer to the general public with the help of exhibits in the Überseemuseum and a lecture in Hall 7.

Memberships
Projects: hyBit, MMM, LogDynamics, π³

hyBit: Hydrogen is seen as the energy carrier of tomorrow. But the path to an efficient hydrogen economy is complex and requires excellent scientific research. The large-scale hydrogen research project hyBit, coordinated by the university, is now receiving nearly 30 million euros in funding for this purpose. Our group is also a member of the project!

MMM: Minds, Media, Machines is an interdisciplinary network of researchers at the University of Bremen and affiliated institutes within and outside the university. They work together to improve our understanding of intelligence and cognition in autonomous agents, agent teams in the context of mediatized worlds. Our research group is also a part of it.

LogDynamics: Since 1995, the Bremen Research Cluster for Dynamics in Logistics (LogDynamics) has stood for interdisciplinary research in logistics topics and for a correspondingly focused structured international doctoral program at the University of Bremen. LogDynamics contributes to the profile of the excellent and modern University of Bremen in research, transfer and doctoral education in an international environment. Through the associated emergences and synergies, LogDynamics provides Bremen with new impulses and driving forces as a business location.

π³: In the graduate program "π³: Parameter Identification - Analysis, Algorithms, Implementations", PhD students concentrate on the interface between applied mathematics and computer science on questions of parameter identification, which are essentially formulated as high-dimensional minimization problems for suitable functionals.

Committee: STACS 2023 PC, WINE 2022 PC

Daniel Schmand was invited to join the program committee of STACS 2023 - Symposium on Theoretical Aspects of Computer Science and has accepted the invitation.

Daniel Schmand was invited to join the program committee of WINE 2022 - The 18th Conference on Web and Internet Economics and has accepted the invitation.