Daniel@0: Daniel@0: Daniel@0: Daniel@0: Netlab Reference Manual gbayes Daniel@0: Daniel@0: Daniel@0: Daniel@0:

gbayes Daniel@0:

Daniel@0:

Daniel@0: Purpose Daniel@0:

Daniel@0: Evaluate gradient of Bayesian error function for network. Daniel@0: Daniel@0:

Daniel@0: Synopsis Daniel@0:

Daniel@0:

Daniel@0: g = gbayes(net, gdata)
Daniel@0: [g, gdata, gprior] = gbayes(net, gdata)
Daniel@0:

Daniel@0: Daniel@0: Daniel@0:

Daniel@0: Description Daniel@0:

Daniel@0: g = gbayes(net, gdata) takes a network data structure net together Daniel@0: the data contribution to the error gradient Daniel@0: for a set of inputs and targets. Daniel@0: It returns the regularised error gradient using any zero mean Gaussian priors Daniel@0: on the weights defined in Daniel@0: net. In addition, if a mask is defined in net, then Daniel@0: the entries in g that correspond to weights with a 0 in the Daniel@0: mask are removed. Daniel@0: Daniel@0:

[g, gdata, gprior] = gbayes(net, gdata) additionally returns the Daniel@0: data and prior components of the error. Daniel@0: Daniel@0:

Daniel@0: See Also Daniel@0:

Daniel@0: errbayes, glmgrad, mlpgrad, rbfgrad

Daniel@0: Pages: Daniel@0: Index Daniel@0:

Daniel@0: