Optimization Over Zonotopes and Training Support Vector Machines

  title={Optimization Over Zonotopes and Training Support Vector Machines},
  author={Marshall W. Bern and David Eppstein},
We make a connection between classical polytopes called zonotopes and Support Vector Machine (SVM) classifiers. We combine this connection with the ellipsoid method to give some new theoretical results on training SVMs. We also describe some special properties of C -SVMs for C → ∞. 
Highly Cited
This paper has 29 citations. REVIEW CITATIONS

From This Paper

Topics from this paper.


Publications citing this paper.


Publications referenced by this paper.
Showing 1-10 of 24 references

A geometri interpretation of -SVM lassi ers

D. J. Crisp, C.J.C. Burges
Advan es in Neural Information Pro essing Systems 12 . S.A. Solla, T.K. Leen, and K.-R. M uller, eds. MIT Press, • 1999
View 5 Excerpts
Highly Influenced

Support-vector networks

Machine Learning • 1995
View 8 Excerpts
Highly Influenced

A fast iterative nearest point algorithm for support ve tor ma hine lassi er design

S. S. Keerthi, S. K. Shevade, C. Bhatta haryya, K.R.K. Murthy
IEEE Trans. Neural Networks • 2000

Polynomial solvability of convex quadratic programming

S. P. Tarasov, L. G. Khachiyan
Soviet Math . Doklady • 2000

Support Ve tor Ma hines

N. Cristianini, J. Shawe-Taylor
Cambridge U. Press, • 2000
View 3 Excerpts

oth. Point sets with many k -sets

Pro . 16th Annual ACM Symp. Computa tional Geometry , • 2000

Input spa e vs. feature spa e in kernel-based methods

B. S holkopf, S. Mika, +4 authors A. J. Smola
IEEE Trans. on Neu ral Networks • 1999

Similar Papers

Loading similar papers…