Timeline for Tikhonov regularization in the non-negative least square - NNLS (python:scipy)
Current License: CC BY-SA 4.0
8 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jun 16, 2020 at 15:20 | comment | added | Brian Borchers | @tBuLi fixed. Thanks for spotting that. | |
| Jun 16, 2020 at 15:19 | history | edited | Brian Borchers | CC BY-SA 4.0 |
added 8 characters in body
|
| Jun 16, 2020 at 15:12 | comment | added | tBuLi | Excellent answer. I do think there is one small mistake: in case you include a prior you also have to multiply that by $\lambda$, so the last entry in $d$ should be $\lambda L x_0$. | |
| Feb 4, 2014 at 2:37 | comment | added | Brian Borchers | There are lots of methods for selecting the regularization parameter. If you know the noise level in $b$, then you can use it as a basis for selecting $\lambda$ (pick the largest lambda that still results in statistically adequate fit to the data.) A simple heuristic that is commonly used in practice is the L-curve criterion- plot $\| Ax - -b \|$ vs. $\| x \|$, and look for a value of $\lambda$ that gives a "corner" solution that is pareto optimal. In practice, the choice of $\lambda$ is often simply subjective- what makes the solution look good. | |
| Feb 4, 2014 at 1:13 | comment | added | user3259573 | Thanks Brian. I implemented that, and it runs. I have one more question: How do I choose the value of lambda? | |
| Feb 1, 2014 at 19:15 | history | edited | Brian Borchers | CC BY-SA 3.0 |
added details to the extension at the end of the asnwer.
|
| Feb 1, 2014 at 16:45 | comment | added | k20 | Nice +1 I would upvote but I'm too lazy to register. | |
| Feb 1, 2014 at 16:31 | history | answered | Brian Borchers | CC BY-SA 3.0 |