A framework is presented for modeling the input-output map f: R/sup N/ to R/sup M/ of artificial neural networks used in pattern classification. Assuming only that f belongs to an appropriate space F (called a neural space) of nonlinear analytic maps, its structure is obtained by requiring that f match a set of exemplary input-output pairs while minimizing a maximum error in an uncertainty ball in F. This criterion guarantees robustness of the solution. The optimal model f thus obtained consists of a two-layer feedforward net, the first layer being the same as the matching score layer of the Hamming net, and the second layer being a new layer (absent in the Hamming net) possessing synaptic weights which result from the described optimization procedure.<<ETX>>
[1]
R. Lippmann,et al.
An introduction to computing with neural nets
,
1987,
IEEE ASSP Magazine.
[2]
D. Luenberger.
Optimization by Vector Space Methods
,
1968
.
[3]
Rui J. P. de Figueiredo,et al.
A best approximation framework and implementation for simulation of large-scale nonlinear systems
,
1980
.
[4]
R. D. Figueiredo.
A generalized Fock space framework for nonlinear system and signal analysis
,
1983
.
[5]
S. Grossberg,et al.
Neural Dynamics of Category Learning and Recognition: Attention, Memory Consolidation, and Amnesia
,
1987
.