Abstract
The field of evolutionary computation has drawn inspiration from Darwinian evolution in which species adapt to the environment through random variations and selection of the fittest. This type of evolutionary computation has found wide applications, but suffers from low efficiency. A recently proposed non- Darwinian form, called Learnable Evolution Model or LEM, applies a learning process to guide evolutionary processes. Instead of random mutations and recombinations, LEM performs hypothesis formation and instantiation. Experiments have shown that LEM may speed-up an evolution process by two or more orders of magnitude over Darwinian-type algorithms in terms of the number of births (or fitness evaluations). The price is a higher complexity of hypothesis formation and instantiation over mutation and recombination operators. LEM appears to be particularly advantageous in problem domains in which fitness evaluation is costly or time-consuming, such as evolutionary design, complex optimization problems, fluid dynamics, evolvable hardware, drug design, and others.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Anderson R.W. (1997), The Baldwin Effect, in Handbook of Evolutionary Computation, Section C3.4.1, pp. C3.:4:1–C3.4:15., IOP Publishing Ltd and Oxford: Oxford University Press.
Baeck, T., Fogel, D.B. and Michalewicz, Z. (eds.) (1997), Handbook of Evolutionary Computation. IOP Publishing Ltd and Oxford: Oxford University Press.
Baldwin, J.M. (1896), A New Factor in Evolution, American Naturalist, vol. 30, pp.441–51.
Banzhaf, W., Nordin P., Keller R.E., and Francone F.D (1998), Genetic Programming: An Introduction, Morgan Kaufman Publishers, Inc., San Francisco, CA, 1998.
Cervone, G., Michalski, R.S., Kaufman K., Panait, L. (2000), Combining Machine Learning with Evolutionary Computation: Recent Results on LEM, Proceedings of the Fifth International Workshop on Multistrategy Learning (MSL2000), Michalski, R.S. and Brazdil, P. B. (eds.), Guimaraes, Portugal, June 5–7.
Cervone, G., Kaufman, K.A., and Michalski, R.S. (2000), Experimental Validations of the Learnable Evolution Model, Proceedings of the 2000 Congress on Evolutionary Computation, La Jolla, California.
Coletti, M., Lash, T., Mandsager, C., Michalski, R.S., and Moustafa, R. (1999), Comparing Performance of the Learnable Evolution Model and Genetic Algorithms on Problems in Digital Signal Filter Design. Proceedings of the 1999 Genetic and Evolutionary Computation Conference (GECCO).
Darwin, C. (1859), On the Origin of Species by Means of Natural Selection, or the Preservation of Favoured Races in the Struggle for Life, John Murray, London.
Goldberg, D.E. (1989), Genetic Algorithms in Search, Optimization and Machine Learning. Addison-Wesley.
Holland, J. (1975), Adaptation in Artificial and Natural Systems. Ann Arbor: The University of Michigan Press.
Kaufman, K. A. and Michalski, R.S. (2000a), Applying Learnable Evolution Model to Heat Exchanger Design, Proceedings of the Seventeenth National Conference on Artificial Intelligence and Twelfth Conference on Innovative Applications of Artificial Intelligence (AAAI-2000/IAAI-2000), Austin, Texas.
Kaufman, K.A. and Michalski, R.S. (2000b), The AQ18 Machine Learning and Data Mining System: An Implementation and User’s Guide. Reports of the Machine Learning Laboratory, George Mason University, Fairfax, VA (to appear).
Koza, J.R., Bennett, F. H. III, Andre D., Keane M. A. (1999), Genetic Programming III: Darwinian Invention and Problem Solving, Morgan Kaufmann Publishers, San Francisco, CA.
Michalewicz, Z. (1996), Genetic Algorithms + Data Structures = Evolutionary Programs. Springer Verlag, Third edition.
Michalski, R.S. (1998), Learnable Evolution: Combining Symbolic and Evolutionary Learning. Proceedings of the Fourth International Workshop on Multistrategy Learning (MSL’98), 14–20.
Michalski, R.S. (2000a), LEARNABLE EVOLUTION MODEL: Evolutionary Processes Guided by Machine Learning. Machine Learning 38(1-2).
Michalski, R.S. (2000b), Natural Induction: A Theory and Methodology of the AQ Approach to Machine Learning and Data Mining. Reports of the Machine Learning Laboratory, George Mason University, Fairfax, VA (to appear).
Michalski. R.S. and Zhang, Q. (1999), Initial Experiments with the LEM1 Learnable Evolution Model: An Application to Function Optimization and Evolvable Hardware. Reports of the Machine Learning and Inference Laboratory, MLI 99-4, George Mason University, Fairfax, VA.
Mitchell, M. (1996). An Introduction to Genetic Algorithms. Cambridge, MA: MIT
Press.Back, Thomas, Optimization by Means of Genetic Algorithms, ENCORE.
Steele E. J. and Blanden R. V. (2000), Lamarck and Antibody Genes, Science, Vol. 288, No. 5475, pp. 2318.
Wnek, J., Kaufman, K., Bloedorn, E. and Michalski, R.S. (1995), Inductive Learning System AQ15c: The Method and User’s Guide, Reports of the Machine Learning and Inference Laboratory, MLI 95-4, George Mason University, Fairfax, VA.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2000 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Michalski, R.S. (2000). Learning and Evolution: An Introduction to Non-darwinian Evolutionary Computation. In: RaÅ›, Z.W., Ohsuga, S. (eds) Foundations of Intelligent Systems. ISMIS 2000. Lecture Notes in Computer Science(), vol 1932. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-39963-1_3
Download citation
DOI: https://doi.org/10.1007/3-540-39963-1_3
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-41094-2
Online ISBN: 978-3-540-39963-6
eBook Packages: Computer ScienceComputer Science (R0)