Daniel@0: Daniel@0: Daniel@0: Daniel@0: Netlab Reference Manual mlphdotv Daniel@0: Daniel@0: Daniel@0: Daniel@0:

mlphdotv Daniel@0:

Daniel@0:

Daniel@0: Purpose Daniel@0:

Daniel@0: Evaluate the product of the data Hessian with a vector. Daniel@0: Daniel@0:

Daniel@0: Synopsis Daniel@0:

Daniel@0:
Daniel@0: hdv = mlphdotv(net, x, t, v)
Daniel@0: Daniel@0: Daniel@0:

Daniel@0: Description Daniel@0:

Daniel@0: Daniel@0:

hdv = mlphdotv(net, x, t, v) takes an MLP network data structure Daniel@0: net, together with the matrix x of input vectors, the Daniel@0: matrix t of target vectors and an arbitrary row vector v Daniel@0: whose length equals the number of parameters in the network, and Daniel@0: returns the product of the data-dependent contribution to the Hessian Daniel@0: matrix with v. The implementation is based on the R-propagation Daniel@0: algorithm of Pearlmutter. Daniel@0: Daniel@0:

Daniel@0: See Also Daniel@0:

Daniel@0: mlp, mlphess, hesschek
Daniel@0: Pages: Daniel@0: Index Daniel@0:
Daniel@0:

Copyright (c) Ian T Nabney (1996-9) Daniel@0: Daniel@0: Daniel@0: Daniel@0: