Using a Markov network as a surrogate fitness function in a genetic algorithm

Surrogate models of fitness have been presented as a way of reducing the number of fitness evaluations required by an evolutionary algorithm. This is of particular interest with expensive fitness functions where the cost of building the model is outweighed by the saving of using fewer function evaluations. In this paper we show how a Markov network model can be used as a surrogate fitness function in a genetic algorithm. We demonstrate this applied to a number of well-known benchmark functions and although the results are good in terms of function evaluations the model-building overhead requires a substantially more expensive fitness function to be worthwhile. We move on to describe a fitness function for feature selection in Case-Based Reasoning, which is considerably more expensive than the other benchmark functions we used. We show that for this problem using the surrogate offers a significant decrease in total run time compared to a GA using the true fitness function.

[1]  Yew-Soon Ong,et al.  A study on polynomial regression and Gaussian process global surrogate model in hierarchical surrogate-assisted evolutionary algorithm , 2005, 2005 IEEE Congress on Evolutionary Computation.

[2]  Ryszard S. Michalski,et al.  LEARNABLE EVOLUTION MODEL: Evolutionary Processes Guided by Machine Learning , 2004, Machine Learning.

[3]  Ron Kohavi,et al.  Irrelevant Features and the Subset Selection Problem , 1994, ICML.

[4]  Bernhard Sendhoff,et al.  Generalizing Surrogate-Assisted Evolutionary Computation , 2010, IEEE Transactions on Evolutionary Computation.

[5]  Qingfu Zhang,et al.  Iterated Local Search with Guided Mutation , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[6]  David E. Goldberg,et al.  Hierarchical BOA Solves Ising Spin Glasses and MAXSAT , 2003, GECCO.

[7]  Pedro Larrañaga,et al.  Estimation of Distribution Algorithms , 2002, Genetic Algorithms and Evolutionary Computation.

[8]  David E. Goldberg,et al.  Evaluation relaxation using substructural information and linear estimation , 2006, GECCO '06.

[9]  Sanmay Das,et al.  Filters, Wrappers and a Boosting-Based Hybrid for Feature Selection , 2001, ICML.

[10]  Gilbert Owusu,et al.  An application of EDA and GA to dynamic pricing , 2007, GECCO '07.

[11]  David E. Goldberg,et al.  Fitness Inheritance In Multi-objective Optimization , 2002, GECCO.

[12]  J. Laurie Snell,et al.  Markov Random Fields and Their Applications , 1980 .

[13]  Vojtech Franc,et al.  Estimation of fitness landscape contours in EAs , 2007, GECCO '07.

[14]  J. A. Lozano,et al.  Estimation of Distribution Algorithms: A New Tool for Evolutionary Computation , 2001 .

[15]  Ivan Koychev,et al.  Feature Selection and Generalisation for Retrieval of Textual Cases , 2004, ECCBR.

[16]  Martin Pelikan,et al.  Fitness Inheritance in the Bayesian Optimization Algorithm , 2004, GECCO.

[17]  Siddhartha Shakya,et al.  Solving the Ising Spin Glass Problem using a Bivariate EDA based on Markov Random Fields , 2006, 2006 IEEE International Conference on Evolutionary Computation.

[18]  Bernhard Sendhoff,et al.  Reducing Fitness Evaluations Using Clustering Techniques and Neural Network Ensembles , 2004, GECCO.

[19]  Pedro Larrañaga,et al.  Evolutionary computation based on Bayesian classifiers , 2004 .

[20]  Yiming Yang,et al.  A Comparative Study on Feature Selection in Text Categorization , 1997, ICML.

[21]  Hod Lipson,et al.  Coevolution of Fitness Predictors , 2008, IEEE Transactions on Evolutionary Computation.

[22]  Siddhartha Shakya,et al.  Using a Markov network model in a univariate EDA: an empirical cost-benefit analysis , 2005, GECCO '05.

[23]  David E. Goldberg,et al.  Combining competent crossover and mutation operators: a probabilistic model building approach , 2005, GECCO '05.

[24]  Bu-Sung Lee,et al.  Memetic algorithm using multi-surrogates for computationally expensive optimization problems , 2007, Soft Comput..

[25]  Gilbert Owusu,et al.  A fully multivariate DEUM algorithm , 2009, 2009 IEEE Congress on Evolutionary Computation.

[26]  Robert E. Smith,et al.  Fitness inheritance in genetic algorithms , 1995, SAC '95.

[27]  John A. W. McCall,et al.  Solving the MAXSAT problem using a multivariate EDA based on Markov networks , 2007, GECCO '07.

[28]  Xavier Llorà,et al.  Do not match, inherit: fitness surrogates for genetics-based machine learning techniques , 2007, GECCO '07.

[29]  Yaochu Jin,et al.  A comprehensive survey of fitness approximation in evolutionary computation , 2005, Soft Comput..

[30]  Marc Schoenauer,et al.  Surrogate Deterministic Mutation: Preliminary Results , 2001, Artificial Evolution.

[31]  Qingfu Zhang,et al.  Approaches to selection and their effect on fitness modelling in an Estimation of Distribution Algorithm , 2008, 2008 IEEE Congress on Evolutionary Computation (IEEE World Congress on Computational Intelligence).

[32]  Arthur C. Sanderson,et al.  Adaptive Differential Evolution , 2009 .

[33]  J. A. Lozano,et al.  Towards a New Evolutionary Computation: Advances on Estimation of Distribution Algorithms (Studies in Fuzziness and Soft Computing) , 2006 .

[34]  Qingfu Zhang,et al.  Structure learning and optimisation in a Markov-network based estimation of distribution algorithm , 2009, 2009 IEEE Congress on Evolutionary Computation.

[35]  Arthur C. Sanderson,et al.  Surrogate Model-Based Differential Evolution , 2009 .

[36]  Khaled Rasheed,et al.  Comparison Of Methods For Using Reduced Models To Speed Up Design Optimization , 2002, GECCO.

[37]  Kok Wai Wong,et al.  Surrogate-Assisted Evolutionary Optimization Frameworks for High-Fidelity Engineering Design Problems , 2005 .

[38]  David E. Goldberg,et al.  Bayesian Optimization Algorithm: From Single Level to Hierarchy , 2002 .

[39]  Alexander E. I. Brownlee,et al.  Multivariate Markov networks for fitness modelling in an estimation of distribution algorithm , 2009 .

[40]  Siddhartha Shakya,et al.  Optimization by estimation of distribution with DEUM framework based on Markov random fields , 2007, Int. J. Autom. Comput..