Performs a sparse online update step of the Gaussian Process.
ogpstep_sp(indX,kX,K1,K2,gamma,hatE)
ogpstep_sp(indX,kX,K1,K2,gamma,hatE)
- updates the GLOBAL Gaussian
process structure net
based on information about the current input.
This function is called from ogppost
.
In the sparse update step the set of basis vectors is not changed! The current input modifies the only the coefficients of the GP.
net
- the GLOBAL Gaussian Process structure to be modified.
indx
- the index of the current input in the training set (needed to update the TAP/EP representation).
kX
- the covariance function of the basis vector set and the current input K0(BV,x).
K1,K2
- the values of the first and second derivatives of the log-average.
gamma
- the geometric component of the updates (the distance of the current input from the subspace given by the basis vector set.
hatE
- the coordinates of the projection of the current input to the basis vector set.
Copyright (c) Lehel Csató (2001-2004)