Matching pursuit

From Wikipedia, the free encyclopedia

You have new messages (last change).

Matching pursuit is a type of numerical technique which involves finding the most "interesting" possible projections in multidimensional data. It is inspired by Projection pursuit, and projections which deviate more from a Normal distribution are considered to be more interesting.

 This statistics-related article is a stub. You can help Wikipedia by expanding it.

[edit] References

1. S. G. Mallat and Z. Zhang, "Matching Pursuits with Time-Frequency Dictionaries", IEEE Transactions on Signal Processing, December 1993, pp. 3397-3415.

[edit] See also

Retrieved from "http://en.wikipedia.org../../../m/a/t/Matching_pursuit.html"