RNNC

Random Neural Net classifier

    W = RNNC(A,N,S)

Input
A	Input dataset
N	Number of neurons in the hidden layer
S	Standard deviation of weights in an input layer (default: 1)

Output
W	Trained Random Neural Net classifier

Description

W is a feed-forward neural net with one hidden layer of N sigmoid neurons.  The input layer rescales the input features to unit variance; the hidden  layer has normally distributed weights and biases with zero mean and  standard deviation S. The output layer is trained by the dataset A.  Default N is number of objects * 0.2, but not more than 100.

If N and/or S is NaN they are optimised by REGOPTC.

Uses the Mathworks' Neural Network toolbox.

Reference(s)

1. W.F. Schmidt, M.A. Kraaijveld, and R.P.W. Duin, Feed forward neural networks with random weights, Proc. ICPR11, Volume II, 1992, 1-4.
2. G.B. Huang, Q.Y. Zhu, C.K. Siew, Extreme learning machine: theory and applications, Neurocomputing, 70 (1), 2006, 489-501

See also

mappings, datasets, lmnc, bpxnc, neurc, rbnc,