tikhonov regularization solution looks very similiar to NNLS solution.
6 views (last 30 days)
Show older comments
Hello,
I am using Tikhonov regularization with non negativity constraints that i am implementing with the lsqnonneg function. I am also using the L curve function provided by regtools to determine the optimal degree of regularization. However, both of these methods provided a solution that is very similar to the solution obtained with ordinary NNLS without regularization. I am wondering if anyone has encountered a similar problem and has suggestions on possible solutions. I am thinking about using the second derivative operator in lieu of the identity matrix in the Tikhonov as well as implementing a truncated SVD. Thanks,
Mike.
0 Comments
Answers (0)
See Also
Categories
Find more on Sparse Matrices in Help Center and File Exchange
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!