pic
The goal of the thesis is to solve the Support Vector Regression problem through second order methods to achieve fast convergence and good accuracy. A primal, unconstrained and twice differentiable formulation of the problem is used. The Representer theorem allows the use of kernels while operating in the primal.
Candidate: Alessio Sarullo
 Graduated April 2016
image_print
Training Support Vector Machines using second order information