Margin Theory
Margin Theory is a theoretical framework in supervised learning that characterizes the generalization performance of classifiers, particularly support vector machines, by the size of the separation (margin) between classes in feature space, distinguishing itself from other learning theories by prioritizing margin maximization as a principle for model robustness and error bounds.
271
Publications
35.4K
Citations
526
Authors
262
Institutions
Publications per year
2017–2026
60
271
A training algorithm for optimal margin classifiers
Bernhard E. Boser, Isabelle Guyon, Vladimir Vapnik · 1992 · 11.6K citations
Boosting the margin: a new explanation for the effectiveness of voting methods
Peter L. Bartlett, Yoav Freund, Wee Sun Lee et al. · The Annals of Statistics · 1998 · 2.3K citations · Full text
On the algorithmic implementation of multiclass kernel-based vector machines
Koby Crammer, Yoram Singer · 2002 · 1.8K citations
1–5 of 271
526
11
h-index
8
h-index
7
h-index
6
h-index
5
h-index
1–5 of 526
262
Nanjing University
Nanjing, China
11 h-index
21 publications
1.1K citations
Google (United States)
Mountain View, United States
10 h-index
15 publications
666 citations
Hebrew University of Jerusalem
Jerusalem, Israel
9 h-index
17 publications
6.9K citations

Australian National University
Canberra, Australia
9 h-index
19 publications
10.6K citations

Chinese Academy of Sciences
Beijing, China
8 h-index
18 publications
724 citations

Peking University
Beijing, China
7 h-index
19 publications
699 citations
Columbia University
New York, United States
7 h-index
13 publications
661 citations
Princeton University
Princeton, United States
6 h-index
8 publications
484 citations
1–8 of 262
Repository · 29 publications · 1.4K citations
18
h-index
11 publications · 4.2K citations
11
h-index
8 publications · 159 citations
8
h-index
7 publications · 132 citations
7
h-index
5 publications · 79 citations
5
h-index
1–5