Timeline for Projecting onto space of matrices with spectral radius less than one
Current License: CC BY-SA 4.0
12 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 20, 2021 at 9:34 | answer | added | Federico Poloni | timeline score: 3 | |
| Dec 20, 2021 at 8:58 | history | edited | Rodrigo de Azevedo | CC BY-SA 4.0 |
added 34 characters in body
|
| Dec 20, 2021 at 6:20 | history | edited | Turbo |
edited tags
|
|
| Dec 19, 2021 at 23:43 | answer | added | Luca Citi | timeline score: 0 | |
| Aug 24, 2021 at 15:48 | comment | added | fedja | I don't think so either, but why do you insist on a norm at all? It looks like what you are after is just something close enough and the solution that is an approximate minimizer (i.e., minimizer up to a constant factor) should be just as good for you as the true minimizer. Am I wrong? | |
| Aug 24, 2021 at 12:50 | comment | added | CComp | I don't think there is a matrix norm for which that is the solution to the minimization problem in the question's first paragraph. Typical norms such as 1 and infinity do not satisfy it. I may be wrong and there is a norm that does? | |
| Aug 21, 2021 at 1:43 | comment | added | fedja | What is wrong with just dividing $A$ by the spectral radius of $|A|$ if the latter exceeds $1$? | |
| Jul 27, 2021 at 22:11 | history | edited | CComp | CC BY-SA 4.0 |
added 6 characters in body
|
| Jul 27, 2021 at 13:53 | history | edited | CComp | CC BY-SA 4.0 |
added 290 characters in body
|
| Jul 27, 2021 at 13:48 | comment | added | CComp | You may be right, the link I provided deals with a different case. Just edited the question (since not really interested in that part I removed it). Do you think that without the absolute value the scenario would simplify significantly? | |
| Jul 27, 2021 at 13:42 | history | edited | CComp | CC BY-SA 4.0 |
Corrected from comment
|
| Jul 27, 2021 at 13:09 | history | asked | CComp | CC BY-SA 4.0 |