Open Azhag opened 12 years ago
Assuming the covariance is Toeplitz, could estimate its inverse from the Fourier transform of the receptive field.
Would then give a closed-form formula for the Fisher information in all cases.
Assuming the covariance is Toeplitz, could estimate its inverse from the Fourier transform of the receptive field.
Would then give a closed-form formula for the Fisher information in all cases.