Automated regularization and svd inverse

[sheryjoe]

Praful here is for discussion

[sheryjoe]

related to #533

[medakk]

What is this issue about?

[sheryjoe]

This is related to autoselecting starting and ending regularization based on the data. More details in #533

The svd inverse might be already addressed (worth checking). The idea here is to use the SVD decomposition of the covariance matrix to invert it, which boils down to inverting the singular values (diagonal matrix).

[medakk]

The svd inverse might be already addressed (worth checking)

Where do I check? Where is the code for this?

[akenmorris]

Can we move this to 6.3?

[sheryjoe]

I don't think we have reached to a consensus yet on this. so yes we can.