ACCIÓN DE INTERNACIONALIZACIÓN EN DOCTORADO: P.D. EN MATEMÁTICAS ( PROF. GILBERT STRANG)

Estudiantes de Doctorado
PDI de la UPV

*Seminario: The Mathematics of Deep Learning: Linear Algebra, Statistics, and Optimization
- Session 1: Applied linear algebra : Matrices and data
Singular values and singular vectors (principal components) of a matrix Best approximation of A by a matrix of low rank
Sparse solutions to Ax = b by choosing the proper norm (not least squares) Randomized linear algebra for large systems
Graphs and their Laplacians for clustering data

- Session 2: The functions of deep learning and stochastic optimization
The architecture of a successful neural net : width and depth Fully connected layers and convolutional layers
Gradient descent to compute optimal weights from the training data The advantages and drawbacks of stochastic gradient descent Accelerated descent by adding momentum and adaptive terms


*Tipo de actividad: reunión con estudiantes de doctorado / directores de tesis
Título: The Mathematics of Deep Learning: Open problems
Fecha: 17/07/2019
Hora: 16:00 h Lugar:
Aula B2 Ed 7A
Duración estimada: 2h
Resumen de contenido:
In this session we will discuss open mathematical problems as well as applications, which are of current interest to researchers in the field of artificial intelligence and deep learning.

*Tipo de actividad: reunión con directores de tesis por proyectos conjuntos
Título: The Mathematics of Deep Learning: Challenge projects
Fecha: 18/07/2019
Hora: 10:00 h Lugar:
Aula B2 Ed 7A
Duración estimada: 2h
Resumen de contenido:
In this session we will present sorne challenge projects, which rnight yield the participation in joint research projects in the field of artificial intelligence and deep learning.