Different scenarios for survival analysis of evolutionary algorithms

Empirical analysis of evolutionary algorithms (EAs) behavior is usually approached by computing relatively simple descriptive statistics like mean fitness and mean number of evaluations to convergence, or more theoretically sound statistical tests for finding significant differences between algorithms. However, these analyses do not consider situations where the EA failed to finish due to numerical errors or excessive computational time. Furthermore, the ability of an EA to continuously make search improvements is usually overlooked. In this paper we propose the use of the theory from survival analysis for empirically investigating the behavior of EAs, even in situations where not all the experiments finish in a reasonable time. We introduce two scenarios for the application of survival analysis in EAs. Survival trees, a machine learning technique adapted to the survival analysis scenario, are applied to automatically identify combinations of EA parameters with similar effect in the behavior of the algorithm.

[1]  Denis Larocque,et al.  A review of survival trees , 2011 .

[2]  Lionel C. Briand,et al.  A practical guide for using statistical tests to assess randomized algorithms in software engineering , 2011, 2011 33rd International Conference on Software Engineering (ICSE).

[3]  Xia Jiang,et al.  A New Method for Predicting Patient Survivorship Using Efficient Bayesian Network Learning , 2014, Cancer informatics.

[4]  E. Kaplan,et al.  Nonparametric Estimation from Incomplete Observations , 1958 .

[5]  Asanao Shimokawa,et al.  Comparison of Splitting Methods on Survival Tree , 2015, The international journal of biostatistics.

[6]  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.

[7]  Francisco Herrera,et al.  Analyzing convergence performance of evolutionary algorithms: A statistical approach , 2014, Inf. Sci..

[8]  A. E. Eiben,et al.  Efficient relevance estimation and value calibration of evolutionary algorithm parameters , 2007, 2007 IEEE Congress on Evolutionary Computation.

[9]  Juan Julián Merelo Guervós,et al.  Using statistical tools to determine the significance and relative importance of the main parameters of an evolutionary algorithm , 2013, Intell. Data Anal..

[10]  M. Greenwood A Report on the Natural Duration of Cancer. , 1926 .

[11]  N. Breslow Covariance analysis of censored survival data. , 1974, Biometrics.

[12]  Seppo J. Ovaska,et al.  A general framework for statistical performance comparison of evolutionary computation algorithms , 2006, Inf. Sci..

[13]  R. Tibshirani The lasso method for variable selection in the Cox model. , 1997, Statistics in medicine.

[14]  W. Spears,et al.  On the Virtues of Parameterized Uniform Crossover , 1995 .

[15]  John McCall,et al.  Statistical analysis of genetic algorithms and inference about optimal factors , 1998 .

[16]  D. Cox Regression Models and Life-Tables , 1972 .

[17]  E. B. Page Ordered Hypotheses for Multiple Treatments: A Significance Test for Linear Ranks , 1963 .

[18]  Hemant Ishwaran,et al.  Evaluating Random Forests for Survival Analysis using Prediction Error Curves. , 2012, Journal of statistical software.

[19]  Leo Breiman,et al.  Classification and Regression Trees , 1984 .

[20]  Marjan Mernik,et al.  A chess rating system for evolutionary algorithms: A new method for the comparison and ranking of evolutionary algorithms , 2014, Inf. Sci..

[21]  D.,et al.  Regression Models and Life-Tables , 2022 .

[22]  Andreas Ziegler,et al.  Random forests for survival analysis using maximally selected rank statistics , 2016, ArXiv.

[23]  Pedro Larrañaga,et al.  Interactions and dependencies in estimation of distribution algorithms , 2005, 2005 IEEE Congress on Evolutionary Computation.

[24]  Marc Parizeau,et al.  DEAP: evolutionary algorithms made easy , 2012, J. Mach. Learn. Res..

[25]  Ritu Gupta,et al.  Statistical exploratory analysis of genetic algorithms , 2004, IEEE Transactions on Evolutionary Computation.

[26]  Kenneth A. De Jong,et al.  A formal analysis of the role of multi-point crossover in genetic algorithms , 1992, Annals of Mathematics and Artificial Intelligence.

[27]  D. Kleinbaum,et al.  Survival Analysis: A Self-Learning Text. , 1996 .

[28]  Günter Rudolph,et al.  Assessing Basin Identification Methods for Locating Multiple Optima , 2016, Advances in Stochastic and Deterministic Global Optimization.

[29]  Roberto Santana,et al.  Reproducing and learning new algebraic operations on word embeddings using genetic programming , 2017, ArXiv.

[30]  K. Hornik,et al.  Unbiased Recursive Partitioning: A Conditional Inference Framework , 2006 .

[31]  Heike Trautmann,et al.  Detecting Funnel Structures by Means of Exploratory Landscape Analysis , 2015, GECCO.

[32]  B. Efron The Efficiency of Cox's Likelihood Function for Censored Data , 1977 .

[33]  Concha Bielza,et al.  Network measures for information extraction in evolutionary algorithms , 2013, Int. J. Comput. Intell. Syst..

[34]  Gerhard Tutz,et al.  A survival tree method for the analysis of discrete event times in clinical and epidemiological studies , 2016, Statistics in medicine.

[35]  K. Hornik,et al.  Model-Based Recursive Partitioning , 2008 .

[36]  T G Clark,et al.  Survival Analysis Part I: Basic concepts and first analyses , 2003, British Journal of Cancer.