Talk:Nonlinear dimensionality reduction

From Wikipedia, the free encyclopedia
Jump to: navigation, search
WikiProject Robotics (Rated C-class, Mid-importance)
WikiProject icon Nonlinear dimensionality reduction is within the scope of WikiProject Robotics, which aims to build a comprehensive and detailed guide to Robotics on Wikipedia. If you would like to participate, you can choose to edit this article, or visit the project page (Talk), where you can join the project and see a list of open tasks.
C-Class article C  This article has been rated as C-Class on the project's quality scale.
 Mid  This article has been rated as Mid-importance on the project's importance scale.
WikiProject Statistics (Rated C-class, Low-importance)
WikiProject icon

This article is within the scope of the WikiProject Statistics, a collaborative effort to improve the coverage of statistics on Wikipedia. If you would like to participate, please visit the project page or join the discussion.

C-Class article C  This article has been rated as C-Class on the quality scale.
 Low  This article has been rated as Low-importance on the importance scale.

Linear Methods[edit]

Why is there a section for linear methods in an article titled non-linear dimensionality reduction? I think it would make more sense to make a new article for linear-dimensionality reduction, and link to it from here. —Preceding unsigned comment added by Headlessplatter (talkcontribs) 16:58, 13 October 2009 (UTC)

I agree with you. I wanted to move it back to the dimension reduction page, but found that it also needs cleaning up. --Dfalcantara (talk) 22:55, 11 January 2010 (UTC)

'A' example is non-linear?[edit]

I'm a little confused about why the first example shows a "nonlinear" reduction in dimensionality. Aren't rotation and scaling linear transformations? Danielx (talk) 06:59, 27 October 2009 (UTC)

If you read the text carefully (where it says 1024 dimensions), it seems that each "data point" may consist of 1024 values of 0 or 1, and in this case "linear" would be shifts, scales and rotations in this 1024 dimension space. That is, not in the 2 dimensional visual space, which would be a different type of image matching problem. Melcombe (talk) 13:34, 27 October 2009 (UTC)
Thanks, Melcombe. That makes sense.

Multidimensional Scaling[edit]

Multidimensional is referred to in the description of some of the techniques, though it is not listed on it's own. Since it is very popular I'm sure it deserves it's own place in the list. T3kcit (talk) 09:22, 4 June 2011 (UTC)

Completeness of list[edit]

It seems that t-SNE is currently among the best performing methods, so should it not be introduced? — Preceding unsigned comment added by (talk) 10:05, 17 June 2011 (UTC)

Thank you, Editors![edit]

I offer my deep thanks for everyone who's put effort into this article. As a student of machine learning, it's one of the most succinct and useful pages I've found. Well done! p.r.newman (talk) 18:02, 11 April 2013 (UTC)

What is nonlinear dimensionality reduction?[edit]

It takes a long time before the article gets to what nonlinear dimensionality reduction actually is. This should be briefly explained already in the very first sentence, otherwise the reader may have no idea what it is he is actually reading about. —Kri (talk) 23:27, 22 March 2014 (UTC)