Abstract
Many learning paradigms for neural networks are based on the optimization of some measure of performance. The learning behavior of these networks then depends on the topography of the corresponding “fitness landscape” and on the chosen optimization method. In this paper, we apply different learning strategies to a number of Boolean problems and analyze how the topography of the fitness landscape is affected by the complexity of the problem and by the network architecture. We further discuss the application of stochastic optimization strategies to problems with delayed performance estimation and present results for a neurocontrol solution to the “lunar lander” problem.
Preview
Unable to display preview. Download preview PDF.
References
G.E. Hinton, Artificial Intelligence 40, 185 (1989).
D.E. Rumelhart, G.E. Hinton, and J.R. Williams, Nature 323, 533 (1986).
“Parallel Distributed Processing: Explorations in the Microstructure of Cognition”, vol. 1: “Foundations”, ed. by D. E. Rumelhart and J. L. McClelland (MIT Press, Cambridge, MA, 1986).
D. H. Ackley, G. E. Hinton, and T. J. Sejnowski, Cognitive Science 9, 147 (1985).
D.J. Montana and L. Davis, in “Proc. 11th Int. Joint Conf. on AI” (Morgan Kaufmann, San Mateo, CA, 1989), pp. 762–767.
H. Mühlenbein and J. Kindermann, in “Connectionism in Perspective”, ed by R. Pfeifer, Z. Schreter, F. Fogelman-Soulie, and L. Steels (North-Holland, Amsterdam, 1989), pp. 173–198.
N. Baba, Neural Networks 2, 367 (1989).
H. Köhler, S. Diederich, W. Kinzel, and M. Opper, Z. Phys. B 78, 333 (1990).
P.J. Werbos, Neural Networks 3, 179 (1990); and references therein.
D. Nguyen and B. Widrow, in “Proc. IJCNN 1989“ (IEEE Service Center, 1989), pp. II-357–II-363.
A.G. Barto, R.S. Sutton, and C.W. Anderson, IEEE Trans. on Systems, Man, and Cybernetics SMC-13, 834 (1983).
B. Widrow and J.M. McCool, IEEE Transactions on Antennas and Propagation, AP-24, 615 (1976).
F.J. Solis and R.J.B. Wets, Mathematics of Operations Research 6, 19 (1981).
S. Kirkpatrick, C.D. Gelatt Jr., and M.P. Vecchi, Science 220, 671 (1983).
R.H.J.M. Otten and L.P.P.P. van Ginneken, “The Annealing Algorithm“ (Kluwer, Boston, 1989).
J. Bernasconi, in “Chaos and Complexity”, ed. by R. Livi, S. Ruffo, S. Ciliberto, and M. Buiatti (World Scientific, Singapore, 1988), pp. 245–259.
L.B. Booker, D.E. Goldberg, and J.H. Holland, Artifical Intelligence 40, 235 (1989).
P. Carnevali and S. Patarnello, Europhys. Lett. 4, 1199 (1987).
N. Tishby, E. Levin, and S.A. Solla, in “Proc. IJCNN 1989“ (IEEE Service Center, 1989), pp. II-403–II-409.
R. Bassein, The American Mathematical Monthly 96, 721 (1989).
Author information
Authors and Affiliations
Editor information
Rights and permissions
Copyright information
© 1990 Springer-Verlag
About this paper
Cite this paper
Bernasconi, J. (1990). Learning and optimization. In: Garrido, L. (eds) Statistical Mechanics of Neural Networks. Lecture Notes in Physics, vol 368. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3540532676_45
Download citation
DOI: https://doi.org/10.1007/3540532676_45
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-53267-5
Online ISBN: 978-3-540-46808-0
eBook Packages: Springer Book Archive