Stagnation Detection with Randomized Local Search

Recently a mechanism called stagnation detection was proposed that automatically adjusts the mutation rate of evolutionary algorithms when they encounter local optima. The so-called SD-(1+1) EA introduced by Rajabi and Witt (GECCO 2020) adds stagnation detection to the classical (1+1) EA with standard bit mutation, which flips each bit independently with some mutation rate, and raises the mutation rate when the algorithm is likely to have encountered local optima. In this paper, we investigate stagnation detection in the context of the k-bit flip operator of randomized local search that flips k bits chosen uniformly at random and let stagnation detection adjust the parameter k. We obtain improved runtime results compared to the SD-(1+1) EA amounting to a speed-up of up to e = 2.71 . . . Moreover, we propose additional schemes that prevent infinite optimization times even if the algorithm misses a working choice of k due to unlucky events. Finally, we present an example where standard bit mutation still outperforms the local k-bit flip with stagnation detection.

[1]  Benjamin Doerr,et al.  Fast genetic algorithms , 2017, GECCO.

[2]  Mahmoud Fouz,et al.  Quasirandom evolutionary algorithms , 2010, GECCO '10.

[3]  Thomas Jansen,et al.  Comparing global and local mutations on bit strings , 2008, GECCO '08.

[4]  Benjamin Doerr,et al.  Theory of Parameter Control for Discrete Black-Box Optimization: Provable Performance Gains Through Dynamic Parameter Choices , 2018, Theory of Evolutionary Computation.

[5]  Benjamin Doerr,et al.  The ($$1+\lambda $$1+λ) Evolutionary Algorithm with Self-Adjusting Mutation Rate , 2018, Algorithmica.

[6]  Frank Neumann,et al.  Randomized Local Search, Evolutionary Algorithms, and the Minimum Spanning Tree Problem , 2004, GECCO.

[7]  Maxim Buzdalov,et al.  The 1/5-th rule with rollbacks: on self-adjustment of the population size in the (1 + (λ, λ)) GA , 2019, GECCO.

[8]  Ingo Wegener,et al.  Methods for the Analysis of Evolutionary Algorithms on Pseudo-Boolean Functions , 2003 .

[9]  Ingo Wegener,et al.  Randomized local search, evolutionary algorithms, and the minimum spanning tree problem , 2004, Theor. Comput. Sci..

[10]  Pierre Hansen,et al.  Variable Neighborhood Search , 2018, Handbook of Heuristics.

[11]  Benjamin Doerr,et al.  k-Bit Mutation with Self-Adjusting k Outperforms Standard Bit Mutation , 2016, PPSN.

[12]  Carsten Witt,et al.  Revised analysis of the (1+1) ea for the minimum spanning tree problem , 2014, GECCO.

[13]  John A. Warwicker,et al.  On the runtime analysis of selection hyper-heuristics for pseudo-Boolean optimisation , 2019 .

[14]  Bryant A. Julstrom,et al.  Biased mutation operators for subgraph-selection problems , 2006, IEEE Transactions on Evolutionary Computation.

[15]  Carsten Witt,et al.  Self-Adjusting Evolutionary Algorithms for Multimodal Optimization , 2020, Algorithmica.

[16]  Benjamin Doerr,et al.  The Impact of Random Initialization on the Runtime of Randomized Search Heuristics , 2015, Algorithmica.

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

[18]  Andrei Lissovoi,et al.  On the Time Complexity of Algorithm Selection Hyper-Heuristics for Multimodal Optimisation , 2019, AAAI.

[19]  Benjamin Doerr,et al.  The (1+λ) evolutionary algorithm with self-adjusting mutation rate , 2017, GECCO.

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

[21]  Pietro Simone Oliveto,et al.  Fast Artificial Immune Systems , 2018, PPSN.

[22]  Benjamin Doerr,et al.  Optimal Static and Self-Adjusting Parameter Choices for the (1+(λ,λ))\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$( , 2017, Algorithmica.

[23]  Per Kristian Lehre,et al.  Drift analysis , 2012, GECCO '12.