VL_NNNOFFSET - CNN norm-dependent offset.

Y = VL_NNNOFFSET(X, PARAM) subtracts from each element of X the weighted norm of the feature channels:

  X(i,j,k) = X(i,j,k) - PARAM(1) * L(i,j) ^ PARAM(2)

where

  L(i,j) = sum_K X(i,j,k)^2

DZDX = VL_NNNOFFSET(X, PARAM, DZDY) computes the derivative of the block projected onto DZDY. DZDX and DZDY have the same dimensions as X and Y respectively.