Seminar: Towards Engineering Stochastic Local Search Algorithms By: Thomas Stuetzle http://iridia.ulb.ac.be/~stuetzle/ Universite Libre de Bruxelles Belgium On: Friday 18th of January 2008 at 10:30 in room 1211 of www.MIC.uu.se Abstract: Stochastic local search (SLS) algorithms are among the most powerful techniques for solving computationally hard problems in many areas of computing science, operations research and engineering. SLS techniques range from rather simple constructive and iterative improvement algorithms to general-purpose SLS methods such as ant colony optimization, iterated local search, memetic algorithms and tabu search. In recent years, it has become evident that the development of effective SLS algorithms is a complex engineering process that typically combines aspects of algorithm design and implementation with empirical analysis and problem-specific background knowledge. This development process needs to be assisted by a sound methodology that addresses the issues arising in the phases of algorithm design, implementation, tuning and experimental evaluation. In this talk, we first give a concise introduction to stochastic local search and present an overview of important topics in the path towards an engineering methodology for stochastic local search algorithms. We will further illustrate several selected topics important for such an engineering methodology using examples from our researches. We end this talk by a discussion of relevant future research topics in SLS.