Hi,
Is there a way to obtain the "Radom matrix" ?
I will try to be more clear.
Based on the direction the signal (the electromagnetic wave), we could define a matrix
of zeros and ones where each row would represent where the signal traveled through the image.
So that we could write something like A * f, where f would be representative of the image, say vec(Image).
Here we can think of the first row of A[1, :] as a row-vector whose components that are 1 represent where the signal
traveled through the image.
More precisely, check this self-consistent material slide 38.
I been to developing some Gaussian process regression approach computing things in a recursive way without
any matrix inversion and would be of great help if I could recover such matrix A from this package.
Any help on this ?
Thank you.