Servizi per la didattica
PORTALE DELLA DIDATTICA

Spectral and machine learning methods for uncertainty quantification

01UJCRV

A.A. 2019/20

Course Language

Inglese

Course degree

Doctorate Research in Electrical, Electronics And Communications Engineering - Torino

Course structure
Teaching Hours
Lezioni 21
Teachers
Teacher Status SSD h.Les h.Ex h.Lab h.Tut Years teaching
Trinchero Riccardo   Ricercatore a tempo det. L.240/10 art.24-B ING-IND/31 9 0 0 0 1
Teaching assistant
Espandi

Context
SSD CFU Activities Area context
*** N/A ***    
2019/20
PERIOD: MARCH - APRIL This course will address the application of two classes of techniques for the uncertainty quantification, that became subject of intensive research in the last decade, namely, spectral methods based on the so-called “polynomial chaos” framework, and Machine Learning techniques, such as Support Vector Machines and Gaussian Process regression. The lessons will cover the main theoretical notions and implementational details. A rigorous mathematical framework is complemented by illustrative examples. At the end of the course, the attendees are expected to be acquainted with the tools and capable of independently apply them to relevant problems in their respective fields of application.
PERIOD: MARCH - APRIL This course will address the application of two classes of techniques for the uncertainty quantification, that became subject of intensive research in the last decade, namely, spectral methods based on the so-called “polynomial chaos” framework, and Machine Learning techniques, such as Support Vector Machines and Gaussian Process regression. The lessons will cover the main theoretical notions and implementational details. A rigorous mathematical framework is complemented by illustrative examples. At the end of the course, the attendees are expected to be acquainted with the tools and capable of independently apply them to relevant problems in their respective fields of application.
Introduction (motivations and definitions). Monte Carlo method. The polynomial chaos expansion and orthogonal polynomials. Stochastic Galerkin method. Stochastic collocation method. Metamodels. Least-square regression method. Stochastic testing method. Global sensitivity analysis. Correlated random parameters. Support Vector Machine and Least-Squares Support Vector Machine regression. Deterministic vs. probabilistic models. Gaussian process regression.
Introduction (motivations and definitions). Monte Carlo method. The polynomial chaos expansion and orthogonal polynomials. Stochastic Galerkin method. Stochastic collocation method. Metamodels. Least-square regression method. Stochastic testing method. Global sensitivity analysis. Correlated random parameters. Support Vector Machine and Least-Squares Support Vector Machine regression. Deterministic vs. probabilistic models. Gaussian process regression.
Modalità di esame:
Exam:


© Politecnico di Torino
Corso Duca degli Abruzzi, 24 - 10129 Torino, ITALY
m@il