The Fast Evaluation Strategy for Evolvable Hardware

An evolutionary algorithm implemented in hardware is expected to operate much faster than the equivalent software implementation. However, this may not be true for slow fitness evaluation applications. This paper introduces a fast evolutionary algorithm (FEA) that does not evaluate all new individuals, thus operating faster for slow fitness evaluation applications. Results of a hardware implementation of this algorithm are presented that show the real time advantages of such systems for slow fitness evaluation applications. Results are presented for six optimisation functions and for image compression hardware.

[1]  John H. Holland,et al.  Adaptation in Natural and Artificial Systems: An Introductory Analysis with Applications to Biology, Control, and Artificial Intelligence , 1992 .

[2]  Mehrdad Salami,et al.  On-line compression of high precision printer images by evolvable hardware , 1998, Proceedings DCC '98 Data Compression Conference (Cat. No.98TB100225).

[3]  Mehrdad Salami,et al.  A Fitness Estimation Strategy for Genetic Algorithms , 2002, IEA/AIE.

[4]  Zbigniew Michalewicz,et al.  Parameter Control in Evolutionary Algorithms , 2007, Parameter Setting in Evolutionary Algorithms.

[5]  Andy J. Keane,et al.  Using Self Organizing Maps and Genetic Algorithms for Model Selection in Multilevel Optimization , 1999, IEA/AIE.

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

[7]  D.E. Goldberg,et al.  Classifier Systems and Genetic Algorithms , 1989, Artif. Intell..

[8]  Zbigniew Michalewicz,et al.  Handbook of Evolutionary Computation , 1997 .

[9]  Mehrdad Salami.  Genetic Algorithm Processor on Reprogrammable Architectures , 1996, Evolutionary Programming.

[10]  D. Goldberg,et al.  Don't evaluate, inherit , 2001 .

[11]  Tetsuya Higuchi,et al.  Evolvable hardware chips for industrial applications , 1999, CACM.

[12]  Guillermo Sapiro,et al.  The LOCO-I lossless image compression algorithm: principles and standardization into JPEG-LS , 2000, IEEE Trans. Image Process..

[13]  Richard J. Carter,et al.  A High-Performance, Pipelined, FPGA-Based Genetic Algorithm Machine , 2001, Genetic Programming and Evolvable Machines.

[14]  M. J. D. Powell,et al.  Direct search algorithms for optimization calculations , 1998, Acta Numerica.

[15]  L. Darrell Whitley,et al.  Island Model genetic Algorithms and Linearly Separable Problems , 1997, Evolutionary Computing, AISB Workshop.

[16]  Mehrdad Salami,et al.  A fast evaluation strategy for evolutionary algorithms , 2003, Appl. Soft Comput..

[17]  Sankar K. Pal,et al.  Genetic Algorithms for Pattern Recognition , 2017 .

[18]  Lothar Thiele,et al.  A Comparison of Selection Schemes Used in Evolutionary Algorithms , 1996, Evolutionary Computation.

[19]  L. Darrell Whitley,et al.  Searching in the Presence of Noise , 1996, PPSN.

[20]  A. Ratle.  Optimal sampling strategies for learning a fitness model , 1999, Proceedings of the 1999 Congress on Evolutionary Computation-CEC99 (Cat. No. 99TH8406).

[21]  Mehrdad Salami,et al.  A Fast Evolutionary Algorithm for Image Compression in Hardware , 2002, IEA/AIE.

[22]  Hans-Paul Schwefel,et al.  Parallel Problem Solving from Nature — PPSN IV , 1996, Lecture Notes in Computer Science.

[23]  Thomas Bäck,et al.  Evolution Strategies on Noisy Functions: How to Improve Convergence Properties , 1994, PPSN.

[24]  Erick Cantú-Paz,et al.  Efficient and Accurate Parallel Genetic Algorithms , 2000, Genetic Algorithms and Evolutionary Computation.

[25]  JesMar ´ in,et al.  Macroevolutionary Algorithms: A New Optimization Method on Fitness Landscapes , 1999 .

[26]  Xin Yao,et al.  Fast Evolution Strategies , 1997, Evolutionary Programming.