Multiplicative Up-Drift

Drift analysis aims at translating the expected progress of an evolutionary algorithm (or more generally, a random process) into a probabilistic guarantee on its run time (hitting time). So far, drift arguments have been successfully employed in the rigorous analysis of evolutionary algorithms, however, only for the situation that the progress is constant or becomes weaker when approaching the target. Motivated by questions like how fast fit individuals take over a population, we analyze random processes exhibiting a (1+δ)\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$(1+\delta )$$\end{document}-multiplicative growth in expectation. We prove a drift theorem translating this expected progress into a hitting time. This drift theorem gives a simple and insightful proof of the level-based theorem first proposed by Lehre (2011). Our version of this theorem has, for the first time, the best-possible near-linear dependence on 1/δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$1/\delta$$\end{document} (the previous results had an at least near-quadratic dependence), and it only requires a population size near-linear in δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\delta$$\end{document} (this was super-quadratic in previous results). These improvements immediately lead to stronger run time guarantees for a number of applications. We also discuss the case of large δ\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\delta$$\end{document} and show stronger results for this setting.

[1]  Weijie Zheng,et al.  Sharp Bounds for Genetic Drift in Estimation of Distribution Algorithms , 2020, IEEE Transactions on Evolutionary Computation.

[2]  Benjamin Doerr,et al.  Probabilistic Tools for the Analysis of Randomized Optimization Heuristics , 2018, Theory of Evolutionary Computation.

[3]  Johannes Lengler,et al.  Drift Analysis , 2017, Theory of Evolutionary Computation.

[4]  Thomas Bäck,et al.  Theory of Evolutionary Computation: Recent Developments in Discrete Optimization , 2020, Theory of Evolutionary Computation.

[5]  Benjamin Doerr,et al.  Sharp Bounds for Genetic Drift in EDAs , 2019, ArXiv.

[6]  Benjamin Doerr,et al.  The efficiency threshold for the offspring population size of the (µ, λ) EA , 2019, GECCO.

[7]  Benjamin Doerr,et al.  Analyzing randomized search heuristics via stochastic domination , 2019, Theor. Comput. Sci..

[8]  P. Lehre,et al.  Level-Based Analysis of the Univariate Marginal Distribution Algorithm , 2018, Algorithmica.

[9]  Benjamin Doerr,et al.  A tight runtime analysis for the (μ + λ) EA , 2018, GECCO.

[10]  Per Kristian Lehre,et al.  University of Birmingham Level-based analysis of the population-based incremental learning algorithm , 2018 .

[11]  Martin S. Krejca,et al.  Intuitive Analyses via Drift Theory , 2018 .

[12]  Martin S. Krejca,et al.  First-Hitting Times Under Additive Drift , 2018, PPSN.

[13]  Benjamin Doerr,et al.  Better Runtime Guarantees via Stochastic Domination , 2018, EvoCOP.

[14]  Benjamin Doerr,et al.  An Elementary Analysis of the Probability That a Binomial Random Variable Exceeds Its Expectation , 2017, Statistics & Probability Letters.

[15]  OneMax,et al.  EA on Generalized Dynamic OneMax , 2018 .

[16]  Benjamin Doerr,et al.  Optimal Static and Self-Adjusting Parameter Choices for the $$(1+(\lambda ,\lambda ))$$(1+(λ,λ)) Genetic Algorithm , 2017, Algorithmica.

[17]  Per Kristian Lehre,et al.  Improved runtime bounds for the univariate marginal distribution algorithm via anti-concentration , 2017, GECCO.

[18]  Duc-Cuong Dang,et al.  Level-Based Analysis of Genetic Algorithms and Other Search Processes , 2014, bioRxiv.

[19]  Duc-Cuong Dang,et al.  Runtime Analysis of Non-elitist Populations: From Classical Optimisation to Partial Information , 2016, Algorithmica.

[20]  Pietro Simone Oliveto,et al.  Improved time complexity analysis of the Simple Genetic Algorithm , 2015, Theor. Comput. Sci..

[21]  Carsten Witt,et al.  (1+1) EA on Generalized Dynamic OneMax , 2015, FOGA.

[22]  Marvin Künnemann,et al.  Optimizing linear functions with the (1+λ) evolutionary algorithm - Different asymptotic runtimes for different instances , 2015, Theor. Comput. Sci..

[23]  Timo Kötzing,et al.  Robustness of Populations in Stochastic Environments , 2014, Algorithmica.

[24]  Timo Kötzing,et al.  Concentration of First Hitting Times Under Additive Drift , 2014, Algorithmica.

[25]  Dirk Sudholt,et al.  The choice of the offspring population size in the (1, λ) evolutionary algorithm , 2014, Theor. Comput. Sci..

[26]  Benjamin Doerr,et al.  Monotonic functions in EC: anything but monotone! , 2014, GECCO.

[27]  Xiequan Fan,et al.  Exponential inequalities for martingales with applications , 2013, 1311.6273.

[28]  Mehryar Mohri,et al.  Tight Lower Bound on the Probability of a Binomial Exceeding its Expectation , 2013, ArXiv.

[29]  Per Kristian Lehre,et al.  Fitness-levels for non-elitist populations , 2011, GECCO '11.

[30]  L. A. Goldberg,et al.  Adaptive Drift Analysis , 2010, Algorithmica.

[31]  Benjamin Doerr,et al.  Multiplicative Drift Analysis , 2010, GECCO '10.

[32]  Daniel Johannsen,et al.  Random combinatorial structures and randomized search heuristics , 2010 .

[33]  Pietro Simone Oliveto,et al.  Theoretical analysis of fitness-proportional selection: landscapes and efficiency , 2009, GECCO.

[34]  Jonathan E. Rowe,et al.  Theoretical analysis of local search strategies to optimize network communication subject to preserving the total number of links , 2009, Int. J. Intell. Comput. Cybern..

[35]  Frank Neumann,et al.  Rigorous analyses of fitness-proportional selection for optimizing linear functions , 2008, GECCO '08.

[36]  Jonathan E. Rowe,et al.  Preliminary theoretical analysis of a local search algorithm to optimize network communication subject to preserving the total number of links , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[37]  Jens Jägersküpper,et al.  Algorithmic analysis of a basic evolutionary algorithm for continuous optimization , 2007, Theor. Comput. Sci..

[38]  Thomas Jansen,et al.  On the brittleness of evolutionary algorithms , 2007, FOGA'07.

[39]  Carsten Witt,et al.  Runtime Analysis of the ( μ +1) EA on Simple Pseudo-Boolean Functions , 2006 .

[40]  Carsten Witt,et al.  Runtime Analysis of the ( + 1) EA on Simple Pseudo-Boolean Functions , 2006, Evolutionary Computation.

[41]  Ali Esmaili,et al.  Probability and Random Processes , 2005, Technometrics.

[42]  Ingo Wegener,et al.  On the Optimization of Monotone Polynomials by Simple Randomized Search Heuristics , 2005, Combinatorics, Probability and Computing.

[43]  Eli Upfal,et al.  Probability and Computing: Randomized Algorithms and Probabilistic Analysis , 2005 .

[44]  Xin Yao,et al.  A study of drift analysis for estimating computation time of evolutionary algorithms , 2004, Natural Computing.

[45]  Ingo Wegener,et al.  Theoretical Aspects of Evolutionary Algorithms , 2001, ICALP.

[46]  Xin Yao,et al.  Drift analysis and average time complexity of evolutionary algorithms , 2001, Artif. Intell..

[47]  G. Grimmett,et al.  Probability and random processes , 2002 .

[48]  D. Freedman On Tail Probabilities for Martingales , 1975 .

[49]  A. Wald On Cumulative Sums of Random Variables , 1944 .