A Gaussian Belief Propagation Solver for Large Scale Support Vector Machines

@article{Bickson2007AGB,
  title={A Gaussian Belief Propagation Solver for Large Scale Support Vector Machines},
  author={Danny Bickson and Elad Yom-Tov and Danny Dolev},
  journal={CoRR},
  year={2007},
  volume={abs/0810.1648}
}
Support vector machines (SVMs) are an extremely successful type of classification and regression algorithms. Building an SVM entails solving a constrained convex quadratic programming problem, which is quadratic in the number of training samples. We introduce an efficient parallel implementation of an SVM solver, based on the Gaussian Belief Propagation algorithm (GaBP). Unlike previous parallel solutions, our approach can be easily used in peer-to-peer and grid environments, where there is no… CONTINUE READING
Highly Cited
This paper has 20 citations. REVIEW CITATIONS