Hauptinhalt

DISCONTINUED
This page is no longer maintained and may be outdated.

---

Dr. Jens Jägersküpper

Address: Informatik 2, TU Dortmund, 44221 Dortmund, Germany Office: Otto-Hahn-Straße 14, room 3.41 E-mail: JJLs2.cs.uni-dortmund.de PGP-key: public key PGP-fingerprint: 14EA 80B5 8D3D 0167 017E 3CFB 129B 4B43 F8F5 0971 Phone / Fax: +49-231-755-2435 / -2047 Support: German Research Foundation (DFG), Collaborative Research Center "Computational Intelligence" (SFB 531)

Research Interests: Algorithms, komplexity, (numerical) optimization; in particular heuristics (theoretical aspects)

Publications

Journals

• Lower bounds for Randomized Direct Search with Isotropic Sampling
  Operations Research Letters 36(3):327-332, 2008.
  A preliminary version (technical report CI-233/07) can be found here.

• Analysis of a simple evolutionary algorithm for minimization in Euclidean spaces
  Theoretical Computer Science 379(3):329-347 (special issue on ICALP 2003), 2007.

• How the (1+1) ES using isotropic mutations minimizes positive definite quadratic forms
  Theoretical Computer Science 361(1):38-56 (special issue on Foundations of GAs 2005), 2006.

Conferences (refereed)

• Oblivious Randomized Direct Search for Real-Parameter Optimization
  Proceedings of the 16th European Symposium on Algorithms (ESA 2008).
  Springer LNCS 5193, pp. 553-564. Available for personal use in PDF  © 2008 Springer-Verlag

• A Blend of Markov-Chain and Drift Analysis
  Proceedings of the 10th Int'l Conference on Parallel Problem Solving from Nature (PPSN 2008).
  Springer LNCS 5199, pp. 41-51. Available for personal use in PDF  © 2008 Springer-Verlag

• together with Mike Preuss:
  Empirical investigation of simplified step-size control in metaheuristics with a view to theory
  Proceedings of the 7th Int'l Workshop on Experimental Algorithms (WEA 2008).
  Springer LNCS 5038, pp. 263-274. Available for personal use in PDF  © 2008 Springer-Verlag

• together with Mike Preuss:
  Aiming for a theoretically tractable CSA variant by means of empirical investigations
  Proceedings of the 2008 Conference on Genetic and Evolutionary Computation (GECCO 2008),
  ACM Press, pp. 503-510. Available for personal use in PDF © ACM, 2008. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proceedings of the 2008 Conference on Genetic and Evolutionary Computation, http://doi.acm.org/10.1145/1389095.1389191.

• Lower bounds for hit-and-run direct search
  Proceedings of the 4th Int'l Symposium on Stochastic Algorithms: Foundations and Applications (SAGA 2007).
  Springer LNCS 4665, pp. 118-129. Available for personal use in PDF  © 2007 Springer-Verlag

• together with Tobias Storch:
  When the Plus Strategy Outperforms the Comma Strategy---and When Not
  Proceedings of the 2007 IEEE Symposium on Foundations of Computational Intelligence (FOCI 2007), IEEE Press, pp. 25-32.

• together with Tobias Storch:
  How Comma Selection Helps with the Escape from Local Optima
  Proceedings of the 2006 Conference on Parallel Problem Solving from Nature (PPSN IX),
  Springer LNCS 4193, pp. 52-61. Available for personal use in PDF  © 2006 Springer-Verlag

• Probabilistic Runtime Analysis of (1+, λ) Evolution Strategies Using Isotropic Mutations
  Proceedings of the 2006 Conference on Genetic and Evolutionary Computation (GECCO 2006),
  ACM Press, pp. 461-468. Available for personal use in PDF © ACM, 2006. This is the author's version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proceedings of the 2006 Conference on Genetic and Evolutionary Computation, http://doi.acm.org/10.1145/1143997.1144083.

• On the Complexity of Overcoming Gaps with Isotropic Mutations and Elitist Selection
  Proceedings of the 2005 IEEE Congress on Evolutionary Computation (CEC 2005), IEEE Press, pp. 206-213.

• together with Carsten Witt:
  Rigorous Runtime Analysis of a (μ+1) ES for the Sphere Function
  Proceedings of the 2005 Conference on Genetic and Evolutionary Computation (GECCO 2005),
  ACM Press, pp. 849-856. Available for personal use in PDF © ACM, 2005. This is the authors' version of the work. It is posted here by permission of ACM for your personal use. Not for redistribution. The definitive version was published in Proceedings of the 2005 Conference on Genetic and Evolutionary Computation, http://doi.acm.org/10.1145/1068009.1068153.

• Rigorous Runtime Analysis of the (1+1) ES: 1/5-Rule and Ellipsoidal Fitness Landscapes
  Foundations of Genetic Algorithms: 8th International Workshop, FoGA 2005, Revised Selected Papers,
  Springer LNCS 3469, pp. 260-281. Available for personal use in PDF  © 2005 Springer-Verlag

• Analysis of a Simple Evolutionary Algorithm for Minimization in Euclidean Spaces
  Proceedings of the 30th International Colloquium on Automata, Languages, and Programming (ICALP 2003)
  Springer LNCS 2719, pp. 1068-1079. Available for personal use in PDF  © 2003 Springer-Verlag

Technical Reports

... can be found here.

Theses

• Probabilistic Analysis of Evolution Strategies Using Isotropic Mutations
  Dissertation (Dr. rer. nat.), University of Dortmund, 2006. Available online here.

• Über den dynamischen (1+1)-EA
  Diploma Thesis (Dipl.-Inform.), University of Dortmund, 2001. Available for personal use in German: PDF.

Misc.

• Slides for teaching convex-hull algorithms (in German) including the algorithms by Jarvis, Graham, and Kirkpatrick/Seidel.
• Slides of my talk at SAGA 2007 on hit-and-run direct search in Euclidean n-space.
• Slides of my (first conference) talk at ICALP 2003 on the runtime analysis of a basic evolutionary algorithm for optimization in Euclidean n-space.

---

Zum Schluss ein kleines Gedicht von P.H. List zum Thema Mittelwert (und dessen Aussagekraft):

Ein Mensch, der von Statistik hört,
denkt dabei nur an Mittelwert.
Er glaubt nicht dran und ist dagegen,
ein Beispiel soll es gleich belegen:

Ein Jäger auf der Entenjagd
hat einen ersten Schuss gewagt.
Der Schuss - zu hastig aus dem Rohr -
lag eine gute Handbreit vor.

Der zweite Schuss mit lautem Krach
lag eine gute Handbreit nach.
Der Jäger spricht ganz unbeschwert
voll Glauben an den Mittelwert:

„Statistisch ist die Ente tot.“

Doch wär' er klug und nähme Schrot
- dies sei gesagt, ihn zu bekehren -
er würde seine Chancen mehren:
Der Schuss geht ab, die Ente stürzt,
weil Streuung ihr das Leben kürzt.