Degree programme Show/Search Programme Course Details Print Save Document

Dettaglio Insegnamento Context Academic Year 2021/2022 Name Dott. - MI (1385) Modelli e Metodi Matematici per l'Ingegneria / Mathematical Models and Methods in Engineering Programme Year 1 Course Details ID Code 056320 Course Title MATHEMATICAL METHODS FOR DEEP LEARNING Course Type MONODISCIPLINARE Credits (CFU / ECTS) 5.0 Course Description The course is split into 4 parts: (a) Introduction to Deep Learning (4h) (b) Approximation Theory and Neural Networks (NN) (6h): old and recent results on the approximation power of NN will be discussed with a particular emphasis on the role of the depth of the NN. (c) Optimization and Deep Neural Networks (10h): Optimization algorithms to train DNN will be presented and discussed with a particular emphasis on nonconvex optimization, recent results dealing with convergence of the learning process and techniques of automatic differentiation. (d) Deep Learning and Partial Differential Equations (6h): the possibility of employing DL in the context of the numerical solution of PDEs will be explored. Scientific-Disciplinary Sector (SSD) SSD Code SSD Description CFU MAT/08 NUMERICAL ANALYSIS 5.0 Details Alphabetical group Professor Course details From (included) To (excluded) A ZZZZ Verani Marco, Miglio Edie

manifestidott v. 1.7.0 / 1.7.0

Area Servizi ICT

23/05/2022