On composing an algorithm portfolio

Evolutionary algorithms are versatile optimization techniques inspired by processes in nature. So far, a wide variety of algorithms have been suggested. However, there is relatively little effort on studying how individual algorithms can work together in a portfolio to achieve a synergy. In this paper, we propose a general methodology to automatically compose a good portfolio from a set of selected evolutionary algorithms. As a single algorithm is a degenerate portfolio, our method also provides an answer to when a portfolio of two or more algorithms are beneficial. Our method has the nice property of being parameter-less; it does not introduce extra parameters. Hence there is no need for parameter control. To illustrate our ideas, we show how a portfolio of five state of the art evolutionary algorithms is automatically constructed using the test functions from the special session on real-parameter optimization of Congress on Evolutionary Computation 2005. It is found that the resulting portfolio obtains the best average ranking. The applicability and limitations of the paradigm of using a benchmarking suite to access evolutionary algorithms are also examined. Though this paper has used evolutionary algorithms only to compose an algorithm portfolio, the idea is generic and is applicable to portfolios with non-evolutionary algorithms as well.

[1]  John Fulcher,et al.  Computational Intelligence: An Introduction , 2008, Computational Intelligence: A Compendium.

[2]  Gary J. Koehler,et al.  Conditions that Obviate the No-Free-Lunch Theorems for Optimization , 2007, INFORMS J. Comput..

[3]  Zbigniew Michalewicz,et al.  Parameter Setting in Evolutionary Algorithms , 2007, Studies in Computational Intelligence.

[4]  Jasper A Vrugt,et al.  Improved evolutionary optimization from genetically adaptive multimethod search , 2007, Proceedings of the National Academy of Sciences.

[5]  Bruce A. Robinson,et al.  Self-Adaptive Multimethod Search for Global Optimization in Real-Parameter Spaces , 2009, IEEE Transactions on Evolutionary Computation.

[6]  H. B. Mann,et al.  On a Test of Whether one of Two Random Variables is Stochastically Larger than the Other , 1947 .

[7]  Antony R. Unwin,et al.  Operations Research — An Introduction (2nd edn) , 1980 .

[8]  David H. Wolpert,et al.  No free lunch theorems for optimization , 1997, IEEE Trans. Evol. Comput..

[9]  Xin Yao,et al.  Population-based Algorithm Portfolios with automated constituent algorithms selection , 2014, Inf. Sci..

[10]  Mehmet Fatih Tasgetiren,et al.  Differential evolution algorithm with ensemble of parameters and mutation strategies , 2011, Appl. Soft Comput..

[11]  Michel Gendreau,et al.  Hyper-heuristics: a survey of the state of the art , 2013, J. Oper. Res. Soc..

[12]  Jason Brownlee,et al.  Clever Algorithms: Nature-Inspired Programming Recipes , 2012 .

[13]  Francisco Herrera,et al.  A study on the use of non-parametric tests for analyzing the evolutionary algorithms’ behaviour: a case study on the CEC’2005 Special Session on Real Parameter Optimization , 2009, J. Heuristics.

[14]  Kenneth Sörensen,et al.  Metaheuristics - the metaphor exposed , 2015, Int. Trans. Oper. Res..

[15]  Shiu Yin Yuen,et al.  Which algorithm should i choose at any point of the search: an evolutionary portfolio approach , 2013, GECCO '13.

[16]  Qingfu Zhang,et al.  Adaptive Operator Selection With Bandits for a Multiobjective Evolutionary Algorithm Based on Decomposition , 2014, IEEE Transactions on Evolutionary Computation.

[17]  Victor O. K. Li,et al.  Real-Coded Chemical Reaction Optimization , 2012, IEEE Transactions on Evolutionary Computation.

[18]  Hamdy A. Taha,et al.  Operations research: an introduction / Hamdy A. Taha , 1982 .

[19]  Yew-Soon Ong,et al.  A Probabilistic Memetic Framework , 2009, IEEE Transactions on Evolutionary Computation.

[20]  Qingfu Zhang,et al.  Differential Evolution With Composite Trial Vector Generation Strategies and Control Parameters , 2011, IEEE Transactions on Evolutionary Computation.

[21]  P. N. Suganthan,et al.  Differential Evolution: A Survey of the State-of-the-Art , 2011, IEEE Transactions on Evolutionary Computation.

[22]  Patrick M. Reed,et al.  Borg: An Auto-Adaptive Many-Objective Evolutionary Computing Framework , 2013, Evolutionary Computation.

[23]  Martin Middendorf,et al.  Performance evaluation of artificial bee colony optimization and new selection schemes , 2011, Memetic Comput..

[24]  James Smith,et al.  A tutorial for competent memetic algorithms: model, taxonomy, and design issues , 2005, IEEE Transactions on Evolutionary Computation.

[25]  Jing J. Liang,et al.  Problem Definitions and Evaluation Criteria for the CEC 2005 Special Session on Real-Parameter Optimization , 2005 .

[26]  Thomas G. Dietterich.  Multiple Classifier Systems , 2000, Lecture Notes in Computer Science.

[27]  Mario A. Muñoz,et al.  The Algorithm Selection Problem on the Continuous Optimization Domain , 2013 .

[28]  Bart Selman,et al.  Algorithm portfolios , 2001, Artif. Intell..

[29]  Bin Li,et al.  Multi-strategy ensemble evolutionary algorithm for dynamic multi-objective optimization , 2010, Memetic Comput..

[30]  Shiu Yin Yuen,et al.  On composing an (evolutionary) algorithm portfolio , 2013, GECCO '13 Companion.

[31]  P. N. Suganthan,et al.  Differential Evolution Algorithm With Strategy Adaptation for Global Numerical Optimization , 2009, IEEE Transactions on Evolutionary Computation.

[32]  Dervis Karaboga,et al.  A powerful and efficient algorithm for numerical function optimization: artificial bee colony (ABC) algorithm , 2007, J. Glob. Optim..

[33]  Andries Petrus Engelbrecht,et al.  Multi-method algorithms: Investigating the entity-to-algorithm allocation problem , 2013, 2013 IEEE Congress on Evolutionary Computation.

[34]  Bin Li,et al.  Multi-strategy ensemble particle swarm optimization for dynamic optimization , 2008, Inf. Sci..

[35]  Fei Peng,et al.  Population-Based Algorithm Portfolios for Numerical Optimization , 2010, IEEE Transactions on Evolutionary Computation.