英文文献蚁群算法.docxVIP

Algorithmica (2009) 54: 243–255 DOI 10.1007/s00453-007-9134-2 Runtime Analysis of a Simple Ant Colony Optimization Algorithm Frank Neumann · Carsten Witt Received: 22 January 2007 / Accepted: 20 November 2007 / Published online: 28 November 2007 ? Springer Science+Business Media, LLC 2007 Abstract Ant Colony Optimization (ACO) has become quite popular in recent years. In contrast to many successful applications, the theoretical foundation of this ran-domized search heuristic is rather weak. Building up such a theory is demanded to understand how these heuristics work as well as to come up with better algorithms for certain problems. Up to now, only convergence results have been achieved show-ing that optimal solutions can be obtained in finite time. We present the first runtime analysis of an ACO algorithm, which transfers many rigorous results with respect to the runtime of a simple evolutionary algorithm to our algorithm. Moreover, we ex-amine the choice of the evaporation factor, a crucial parameter in ACO algorithms, in detail. By deriving new lower bounds on the tails of sums of independent Poisson trials, we determine the effect of the evaporation factor almost completely and prove a phase transition from exponential to polynomial runtime. Keywords Randomized search heuristics · Ant colony optimization · Runtime analysis A preliminary version of this paper appeared in the Proceedings of the 17th International Symposium on Algorithms and Computation (ISAAC 2006), volume 4288 of LNCS, pp. 618–627, Springer. Financial support for C. Witt by the Deutsche Forschungsgemeinschaft (SFB) in terms of the Collaborative Research Center “Computational Intelligence” (SFB 531) is gratefully acknowledged. F. Neumann ( ) Algorithms and Complexity, Max-Planck-Institut für Informatik, 66123 Saarbrücken, Germany e-mail: fne@mpi-inf.mpg.de C. Witt FB Informatik, LS 2, Universit?t Dortmund, 44221 Dortmund, Germany e-mail: carsten.witt@cs.uni-dortmund.de 244 Algorithmica (2009)