Corso di “Mathematics for Machine Learning” – Inizio 21.05.2018, ore 09:30 – Aula Seminari (Piano terra, Cubo 44Z – Ex Presidenza di Ingegneria)

Mathematics for Machine Learning – Inizio corso 21.05.2018, ore 09:30 – Aula Seminari (Piano terra, Cubo 44Z – Ex Presidenza di Ingegneria)

Prof. Manlio Gaudioso

The course is meant to provide basic Numerical Optimization tools to handle some classes of  Machine Learning problems such as supervised and unsupervised classification. In particular notions on optimality conditions for functions of several variables both in the constrained and the unconstrained case will be recalled. The problem of separating sets in n-dimension spaces  by appropriate separation surfaces will be put in the form of an optimization problem. The  Support Vector Machine (SVM)  model, where separating hyperplanes are adopted for classification purposes, will  be discussed, together with possible alternative separation methods based on piecewise affine, ellipsoidal, spherical or conical surfaces.  Multiple instance classification models  will be discussed as well, together with the results of some applications.

Orari:

  • Lunedì 21 Maggio: ore 9.30-11.30;
  • Martedì 22 Maggio: ore 11.00-13.00;
  • Mercoledì 23 Maggio: ore 15.00-17.00;
  • Giovedì  24 Maggio: ore 9.30-11.30;
  • Venerdì 25Maggio: ore 9.30-11.30;