×

Architecture-independent approximation of functions. (English) Zbl 0983.68160

Summary: We show that minimizing the expected error of a feedforward network over a distribution of weights results in an approximation that tends to be independent of network size as the number of hidden units grows. This minimization can be easily performed, and the complexity of the resulting function implemented by the network is regulated by the variance of the weight distribution. For a fixed variance, there is a number of hidden units above which either the implemented function does not change or the change is slight and tends to zero as the size of the network grows. In sum, the control of the complexity depends on only the variance, not the architecture, provided it is large enough.

MSC:

68T05 Learning and adaptive systems in artificial intelligence
PDFBibTeX XMLCite
Full Text: DOI Link

References:

[1] DOI: 10.1162/neco.1996.8.3.643 · Zbl 05477959 · doi:10.1162/neco.1996.8.3.643
[2] DOI: 10.1016/0167-2789(90)90081-Y · doi:10.1016/0167-2789(90)90081-Y
[3] DOI: 10.1016/0893-6080(90)90044-L · doi:10.1016/0893-6080(90)90044-L
[4] DOI: 10.1109/72.238328 · doi:10.1109/72.238328
[5] DOI: 10.1109/72.317730 · doi:10.1109/72.317730
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.