论文信息 - Exact Classification with Two-Layered Perceptrons

Exact Classification with Two-Layered Perceptrons

We study the capabilities of two-layered perceptrons for classifying exactly a given subset. Both necessary and sufficient conditions are derived for subsets to be exactly classifiable with two-layered perceptrons that use the hard-limiting response function. The necessary conditions can be viewed as generalizations of the linear-separability condition of one-layered perceptrons and confirm the conjecture that the capabilities of two-layered perceptrons are more limited than those of three-layered perceptrons. The sufficient conditions show that the capabilities of two-layered perceptrons extend beyond the exact classification of convex subsets. Furthermore, we present an algorithmic approach to the problem of verifying the sufficiency condition for a given subset.

Emile H. L. Aarts | Patrick J. Zwietering | Jaap Wessels

[1] Leong Kwan Li. On Computing Decision Regions with Neural Nets , 1991, J. Comput. Syst. Sci..

[2] Kurt Hornik,et al. Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

[3] Emile H. L. Aarts,et al. The Design and Complexity of Exact Multilayered Perceptrons , 1991, Int. J. Neural Syst..

[4] G. J. Gibson,et al. On the decision regions of multilayer perceptrons , 1990, Proc. IEEE.

[5] Richard Lippmann,et al. Neural Net and Traditional Classifiers , 1987, NIPS.

[6] George Cybenko,et al. Complexity Theory of Neural Networks and Classification Problems , 1990, EURASIP Workshop.

[7] Marvin Minsky,et al. Perceptrons: An Introduction to Computational Geometry , 1969 .

[8] Geoffrey E. Hinton,et al. Learning internal representations by error propagation , 1986 .

[9] R. Lippmann,et al. An introduction to computing with neural nets , 1987, IEEE ASSP Magazine.

[10] Ehl Emile Aarts,et al. Neural networks and production planning , 1991 .

[11] Ken-ichi Funahashi,et al. On the approximate realization of continuous mappings by neural networks , 1989, Neural Networks.

[12] George Cybenko,et al. Approximation by superpositions of a sigmoidal function , 1989, Math. Control. Signals Syst..