Abstract
We introduce multiplicative drift analysis as a suitable way to analyze the runtime of randomized search heuristics such as evolutionary algorithms. Our multiplicative version of the classical drift theorem allows easier analyses in the often encountered situation that the optimization progress is roughly proportional to the current distance to the optimum.
To display the strength of this tool, we regard the classical problem of how the (1+1) Evolutionary Algorithm optimizes an arbitrary linear pseudo-Boolean function. Here, we first give a relatively simple proof for the fact that any linear function is optimized in expected time O(nlogn), where n is the length of the bit string. Afterwards, we show that in fact any such function is optimized in expected time at most (1+o(1))1.39enlnn, again using multiplicative drift analysis. We also prove a corresponding lower bound of (1−o(1))enlnn which actually holds for all functions with a unique global optimum.
We further demonstrate how our drift theorem immediately gives natural proofs (with better constants) for the best known runtime bounds for the (1+1) Evolutionary Algorithm on combinatorial problems like finding minimum spanning trees, shortest paths, or Euler tours in graphs.
Similar content being viewed by others
Notes
By ℕ:={0,1,2,…} we denote the set of non-negative integers and by ℝ we denote the set of real numbers.
We might as well perform our analysis of g as defined in (11). However, our choice of g does not make the somewhat artificial binary distinction between bits with high and low indices and, thus, seems to be more natural.
In [23], Neumann and Wegener actually analyze the MST problem on the larger search space consisting of all subgraphs of G. In this section, we restrict the problem to the sub-problem of finding a MST when starting with an arbitrary spanning tree. It is shown in [23] that this sub-problem is the bottleneck in the runtime analysis of the (1+1) EA optimizing the MST problem.
References
Auger, A., Doerr, B. (eds.): Theory of Randomized Search Heuristics. Series on Theoretical Computer Science, vol. 1. World Scientific, Singapore (2011)
Baswana, S., Biswas, S., Doerr, B., Friedrich, T., Kurur, P., Neumann, F.: Computing single source shortest paths using single-objective fitness. In: FOGA ’09: Proceedings of the 10th ACM Workshop on Foundations of Genetic Algorithms, pp. 59–66. ACM, New York (2009)
Doerr, B., Goldberg, L.A.: Drift analysis with tail bounds. In: PPSN ’10: Proceedings of the 11th International Conference on Parallel Problem Solving from Nature, pp. 174–183. Springer, Berlin (2010)
Doerr, B., Goldberg, L.A.: Adaptive drift analysis. In: PPSN ’10: Proceedings of the 11th International Conference on Parallel Problem Solving from Nature, pp. 32–41. Springer, Berlin (2010)
Doerr, B., Johannsen, D.: Adjacency list matchings—an ideal genotype for cycle covers. In: GECCO ’07: Proceedings of the 9th Annual Genetic and Evolutionary Computation Conference, pp. 1203–1210. ACM, New York (2007)
Doerr, B., Johannsen, D.: Edge-based representation beats vertex-based representation in shortest path problems. In: GECCO ’10: Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, pp. 759–766. ACM, New York (2010)
Doerr, B., Fouz, M., Witt, C.: Quasirandom evolutionary algorithms. In: GECCO ’10: Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, pp. 1457–1464. ACM, New York (2010)
Doerr, B., Jansen, T., Sudholt, D., Winzen, C., Zarges, C.: Optimizing monotone functions can be difficult. In: PPSN ’10: Proceedings of the 11th International Conference on Parallel Problem Solving from Nature, pp. 42–51. Springer, Berlin (2010)
Doerr, B., Johannsen, D., Winzen, C.: Drift analysis and linear functions revisited. In: CEC ’10: Proceedings of the 2010 IEEE Congress on Evolutionary Computation, pp. 1967–1974. IEEE, New York (2010)
Doerr, B., Johannsen, D., Winzen, C.: Multiplicative drift analysis. In: GECCO ’10: Proceedings of the 12th Annual Genetic and Evolutionary Computation Conference, pp. 1449–1456. ACM, New York (2010)
Droste, S., Jansen, T., Wegener, I.: On the analysis of the (1+1) evolutionary algorithm. Theor. Comput. Sci. 276(1–2), 51–81 (2002)
Giel, O., Lehre, P.K.: On the effect of populations in evolutionary multi-objective optimization. In: GECCO ’06: Proceedings of the 8th Annual Genetic and Evolutionary Computation Conference, pp. 651–658. ACM, New York (2006)
Giel, O., Wegener, I.: Evolutionary algorithms and the maximum matching problem. In: STACS ’03: Proceedings of the 20th Annual Symposium on Theoretical Aspects of Computer Science. Lecture Notes in Computer Science, vol. 2607, pp. 415–426. Springer, Berlin (2003)
Hajek, B.: Hitting-time and occupation-time bounds implied by drift analysis with applications. Adv. Appl. Probab. 14(3), 387–403 (1982)
Happ, E., Johannsen, D., Klein, C., Neumann, F.: Rigorous analyses of fitness-proportional selection for optimizing linear functions. In: GECCO ’08: Proceedings of the 10th Annual Genetic and Evolutionary Computation Conference, pp. 953–960. ACM, New York (2008)
He, J., Yao, X.: Drift analysis and average time complexity of evolutionary algorithms. Acta Inform. 127(1), 51–81 (2001)
He, J., Yao, X.: Erratum to: Drift analysis and average time complexity of evolutionary algorithms [Artificial Intelligence 127 (2001) 57–85]. Acta Inform. 140(1–2), 245–248 (2002)
He, J., Yao, X.: A study of drift analysis for estimating computation time of evolutionary algorithms. Nat. Comput. 3(1), 21–35 (2004)
Jägersküpper, J.: A blend of Markov-chain and drift analysis. In: PPSN ’08: Proceedings of the 10th International Conference on Parallel Problem Solving from Nature, pp. 41–51. Springer, Berlin (2008)
Jansen, T., Oliveto, P.S., Zarges, C.: On the analysis of the immune-inspired B-Cell Algorithm for the vertex cover problem. In: ICARIS ’11: Proceedings of the 10th International Conference on Artificial Immune Systems. Lecture Notes in Computer Science, vol. 6825, pp. 117–131. Springer, Berlin (2011)
Kano, M.: Maximum and kth maximal spanning trees of a weighted graph. Combinatorica 7(2), 205–214 (1987)
Mühlenbein, H.: How genetic algorithms really work: mutation and hill-climbing. In: PPSN ’92: Proceedings of the 2nd International Conference on Parallel Problem Solving from Nature, pp. 15–25 (1992)
Neumann, F., Wegener, I.: Randomized local search, evolutionary algorithms, and the minimum spanning tree problem. Theor. Comput. Sci. 378(1), 32–40 (2007)
Neumann, F., Oliveto, P.S., Witt, C.: Theoretical analysis of fitness-proportional selection: landscapes and efficiency. In: GECCO ’09: Proceedings of the 11th Annual Genetic and Evolutionary Computation Conference, pp. 835–842. ACM, New York (2009)
Oliveto, P.S., Witt, C.: Simplified drift analysis for proving lower bounds in evolutionary computation. Algorithmica 59(3), 369–386 (2011)
Scheder, D., Welzl, E.: Private communications (2008)
Witt, C.: Tight bounds on the optimization time of the (1+1) EA on linear functions. In: STACS ’12: Proceedings of the 29th Annual Symposium on Theoretical Aspects of Computer Science (2012, to appear)
Acknowledgements
We like to thank Dirk Sudholt for pointing out that Theorem 11 holds for all pseudo-Boolean function with a unique global optimum rather than only for all linear functions.
Author information
Authors and Affiliations
Corresponding author
Additional information
Daniel Johannsen is supported by a fellowship within the Postdoc-Programme of the German Academic Exchange Service (DAAD).
Carola Winzen is a recipient of the Google Europe Fellowship in Randomized Algorithms, and this work is supported in part by this Google Fellowship.
Rights and permissions
About this article
Cite this article
Doerr, B., Johannsen, D. & Winzen, C. Multiplicative Drift Analysis. Algorithmica 64, 673–697 (2012). https://doi.org/10.1007/s00453-012-9622-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00453-012-9622-x