User talk:Smarchesini

Welcome!

Hello, Smarchesini, and welcome to Wikipedia! Thank you for your contributions. I hope you like the place and decide to stay. Here are some pages that you might find helpful:

I hope you enjoy editing here and being a Wikipedian! Please sign your name on talk pages using four tildes (~~~~); this will automatically produce your name and the date. If you need help, check out Wikipedia:Questions, ask me on my talk page, or ask your question and then place {{helpme}} before the question on your talk page. Again, welcome! – Oleg Alexandrov (talk) 02:41, 15 May 2007 (UTC)[reply]

Secant method

I removed your edit to Secant method, because I don't think it is correct. There is no derivative in the definition of the secant method in 1D, it is just a line between two points on the graph of the function. I don't think your "generalized" method is the secant method, and as such, I don't know if it belongs to the secant method article. Are you referring instead to Newton's method by chance?

You can reply here. Thanks. Oleg Alexandrov (talk) 03:32, 3 August 2007 (UTC)[reply]

My apologies, you are right. Quasi-newton methods however are a generalization of the secant method to multidimensional problems. They are normally used to find the root of the gradient, rather than a function because that is what is useful in optimization.

I corrected the text. —The preceding unsigned comment was added by Smarchesini (talk • contribs).

I am still confused however. By "this iterative scheme" you are referring to which one from above? Is your function

f:\mathbb {R} ^{n}\to \mathbb {R} ^{n}

? If so, the jacobian is well-defined. If not, your equation may not have a unique solution (number of equations does not equal the number of unknowns). I think that text needs more work.

Please reply on this page, to keep all conversation in the same place. Oleg Alexandrov (talk) 14:55, 3 August 2007 (UTC)[reply]

Your recent edits improved that section. But are you sure the new method is called the "Multidimensional Secant method" and that the article secant method is the right place for discussing that? Oleg Alexandrov (talk) 00:14, 5 August 2007 (UTC)[reply]

Thanks for expanding Broyden's method. I replaced the text you put in at secant method with a link to Broyden's method, to avoid repeating the same thing in two places. Feel free to correct that if you don't agree with it. (Note that your text at secant method had a mistake by the way, instead of $J_{n+1}$ it should have had $J_{n}$ I guess). Oleg Alexandrov (talk) 01:06, 6 August 2007 (UTC)[reply]