Sciweavers
Explore
Publications
Books
Software
Tutorials
Presentations
Lectures Notes
Datasets
Labs
Conferences
Community
Upcoming
Conferences
Top Ranked Papers
Most Viewed Conferences
Conferences by Acronym
Conferences by Subject
Conferences by Year
Tools
PDF Tools
Image Tools
Text Tools
OCR Tools
Symbol and Emoji Tools
On-screen Keyboard
Latex Math Equation to Image
Smart IPA Phonetic Keyboard
Community
Sciweavers
About
Terms of Use
Privacy Policy
Cookies
107
click to vote
ICML
2008
IEEE
99
views
Machine Learning
»
more
ICML 2008
»
Random classification noise defeats all convex potential boosters
16 years 4 months ago
Download
www.phillong.info
A broad class of boosting algorithms can be interpreted as performing coordinate-wise gradient descent to minimize some potential function of the margins of a data set. This class...
Philip M. Long, Rocco A. Servedio
claim paper
Read More »