Truncated SVD vs Tikhonov regularization to compute pseudo inverse Cm_inv in beamforming

Dear all,

I recently tested the DICS beamforming and compared results I obtained
using MNE with those I perviously got using Fieldtrip. I came across a
potential bug related to the computation of the inverse cross-spectral
matrix. I created an issue on github:

https://github.com/mne-tools/mne-python/issues/3832

Could you please have a look at it ?

Thanks a lot

Andrea