Abstract
Evolvable Hardware (EHW) is a hardware which modifies its own hardware structure according to the environmental changes. EHW is implemented on a programmable logic device (PLD), whose architecture can be altered by downloading a binary bit string, i.e. architecture bits. The architecture bits are adaptively acquired by genetic algorithms (GA). The target task of EHW is “Boolean concept formation”, which has been intensively studied in machine learning literatures. Although many evolutionary or adaptive techniques were proposed to solve this class of problems, there have been very few comparative studies from the viewpoint of computational learning theory. This paper describes machine learning approach to the gate-level EHW, i.e. 1) MDL-based improvement of fitness evaluation, and 2) comparative studies of efficiency by PAC criterion. We also discuss the current extension of EHW and related works.
This paper presented a machine learning approach to gate-level EHW, i.e.
1. MDL-based improvement of fitness evaluation has been introduced for the sake of the robustness of EHW.
2. Comparative studies have been constructed from a viewpoint of PAC learning theory, in order to evaluate the performance of several adaptive methods.
We believe this is a step toward the integration of the practical field, i.e. EHW, and the theoretical field, i.e. machine learning.
This is a preview of subscription content, log in via an institution to check access.
Preview
Unable to display preview. Download preview PDF.
References
Akaike, H., On the Entropy Maximization Principle, in Applications of Statistics, Krishnaiah, P.R. (ed.), North-Holland, 1977
Anthony, M. and Biggs, N., Computational Learning Theory, Cambridge Tracts in Theoretical Computer Science 30, Cambridge, 1992
Armstrong, W.W. and Gecsei, J., Adaptation Algorithms for Binary Tree Networks, IEEE TR. SMC, SMC-9, No. 5, 1979
Barron, A.R. Minimum Complexity Density Estimation, in IEEE Tr. on Information Theory, vol. 37, no. 4, pp. 1034–1054, 1991
Goldberg, D. et al., Rapid Accurate Optimization of Difficult Problems using Fast Messy Genetic Algorithms, Proc. 5th Int. Joint Conf. on Genetic Algorithms (ICGA93), 1993
Higuchi, T. et al., Evolvable Hardware with Genetic Learning, in Massively Parallel Artificial Intelligence (eds. H. Kitano), MIT Press, pp 398–421, 1994.
Iba, H., deGaris, H. and Sato, T., Genetic Programming using a Minimum Description Length Principle, in Advances in Genetic Programming, (ed. Kenneth E. Kinnear, Jr.), MIT Press, pp. 265–284, 1994
Iba, H., deGaris, H. and Sato, T., Numerical Approach to Genetic Programming for System Identification, in Evolutionary Computation, vol. 3, no. 4, 1996
Iwata, M., Kajitani, I., Yamada, H., Iba, H. and Higuchi, T., A Pattern Recognition System Using Evolvable Hardware, in Proc. of Parallel Problem Solving from Nature IV (Berlin, Germany: Springer-Verlag), 1996
Kajitani, I. et al., Variable Length Chromosome GA for Evolvable Hardware in Proc. of 3rd Int. Conf. on Evolutionary Computation (ICEC96), 1996.
Konagaya, A. and Kondo, H., Stochastic Motif Extraction using a Genetic Algorithm with the MDL Principle, in Hawaii Int. Conf. of Computer Systems, 1993
Murakawa, M., Yoshizawa, S., Kajitani, I., Furuya, T., Iwata, M. and Higuchi, T., Hardware Evolution at Function Levels, Proc. of Parallel Problem Solving from Nature IV (Berlin, Germany: Springer-Verlag), 1996
Pagallo, G. and Hausslear, D. Boolean Feature Discovery in Empirical Learning, Machine Learning, vol. 5, 1990
Rissanen, J., Stochastic Complexity in Statistical Inquiry, World Scientific, 1989
Rumelhart, D.E. and McClelland, J.L. Parallel Distributed Processing, MIT Press, 1986
Seshu, R., Binary Decision Trees and an “Average-case” Model for Concept Learning: Implications for Feature Construction and the Study of Bias, in Computational Learning Theory and Natural Learning Systems, Volume I: Constraints and Prospects, Hanson, J. et al. (eds.), MIT Press, pp. 213–248, 1994
Tenorio, M.F. and Lee, W. Self-organizing Network for Optimum Supervised Learning, IEEE Tr. Neural Networks, vol. 1, no. 1, pp. 100–109, 1990
Wallace, C.S. and Freeman, P.R. Estimation and Inference by Compact Coding, in Journal of Royal Statistical Society, Ser.B, vol. 49, no. 3, pp. 240–265, 1987
Weigend, A.S. and Rumelhart, D.E., Weight Elimination and Effective Network Size, in Computational Learning Theory and Natural Learning Systems, Volume I: Constraints and Prospects, Hanson, J. et al. (eds.), MIT Press, pp. 457–476, 1994
Wilson, S.W., Classifier Systems and the Animat Problem, Machine Learning, vol. 2, no. 3, 1987
Yamanishi, K., and Konagaya, A., “Learning Stochastic Motifs from Genetic Sequences”, in Proc. of 8th International Workshop on Machine Learning, pp. 467–471, 1991
Yamanishi, K., A Learning Criterion for Stochastic Rules, in Machine Learning, vol. 9, pp. 165–204, 1992
Zhang, B.-T. and Mühlenbein, H., Balancing Accuracy and Parsimony in Genetic Programming, in Evolutionary Computation, vol. 3, no. 1, pp. 17–38, 1995
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1997 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Iba, H., Iwata, M., Higuchi, T. (1997). Machine learning approach to gate-level Evolvable Hardware. In: Higuchi, T., Iwata, M., Liu, W. (eds) Evolvable Systems: From Biology to Hardware. ICES 1996. Lecture Notes in Computer Science, vol 1259. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-63173-9_57
Download citation
DOI: https://doi.org/10.1007/3-540-63173-9_57
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-63173-6
Online ISBN: 978-3-540-69204-1
eBook Packages: Springer Book Archive