Using Smart Sampling to Discover Promising Regions and Increase the Efficiency of Differential Evolution

This paper presents a novel method to discover promising regions in a continuous search space. Using machine learning techniques, the algorithm named Smart Sampling was tested in hard known benchmark functions, and was able to find promising regions with solutions very close to the global optimum, significantly decreasing the number of evaluations needed by a metaheuristic to finally find this global optimum, when heuristically started inside a promising region. Results show favorable agreement with theories which state the importance of an adequate starting population. The results also present significant improvement in the efficiency of the tested metaheuristic, without adding any parameter, operator or strategy. Being a technique which can be used by any populational metaheuristic, the work presented here has profound implications for future studies of global optimization and may help solve considerably difficult optimization problems.

[1]  Michael N. Vrahatis,et al.  The New k-Windows Algorithm for Improving the k-Means Clustering Algorithm , 2002, J. Complex..

[2]  Márk Jelasity,et al.  UEGO, an Abstract Clustering Technique for Multimodal Global Optimization , 2001, J. Heuristics.

[3]  Dimitris K. Tasoulis,et al.  Improving the orthogonal range search k-windows algorithm , 2002, 14th IEEE International Conference on Tools with Artificial Intelligence, 2002. (ICTAI 2002). Proceedings..

[4]  Patrick Siarry,et al.  Genetic and Nelder-Mead algorithms hybridized for a more accurate global optimization of continuous multiminima functions , 2003, Eur. J. Oper. Res..

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

[6]  Thomas G. Dietterich.  What is machine learning? , 2020, Archives of Disease in Childhood.

[7]  Shahryar Rahnamayan,et al.  Opposition versus randomness in soft computing techniques , 2008, Appl. Soft Comput..

[8]  Dimitris K. Tasoulis,et al.  Clustering in evolutionary algorithms to efficiently compute simultaneously local and global minima , 2005, 2005 IEEE Congress on Evolutionary Computation.

[9]  Uday K. Chakraborty,et al.  Advances in Differential Evolution , 2010 .

[10]  Rainer Storn,et al.  Differential Evolution – A Simple and Efficient Heuristic for global Optimization over Continuous Spaces , 1997, J. Glob. Optim..

[11]  David W. Aha,et al.  Instance-Based Learning Algorithms , 1991, Machine Learning.

[12]  Hamid R. Tizhoosh,et al.  Opposition-Based Learning: A New Scheme for Machine Intelligence , 2005, International Conference on Computational Intelligence for Modelling, Control and Automation and International Conference on Intelligent Agents, Web Technologies and Internet Commerce (CIMCA-IAWTIC'06).

[13]  Alexandre César Muniz de Oliveira,et al.  Hybrid Evolutionary Algorithms and Clustering Search , 2007 .

[14]  Vinicius Veloso de Melo,et al.  Discovering Promising Regions to Help Global Numerical Optimization Algorithms , 2007, MICAI.

[15]  William W. Cohen.  Fast Effective Rule Induction , 1995, ICML.

[16]  Márk Jelasity,et al.  Reliability and Performance of UEGO, a Clustering-based Global Optimizer , 2001, J. Glob. Optim..

[17]  Kalyanmoy Deb,et al.  An Investigation of Niche and Species Formation in Genetic Function Optimization , 1989, ICGA.