Phelma Formation 2022

Artificial Neural Networks - WPMGRNF7

  • Number of hours

    • Lectures 6.0
    • Projects 0
    • Tutorials 12.0
    • Internship 0
    • Laboratory works 0

    ECTS

    ECTS 2.0

Goal(s)

This course is intended to provide the necessary foundation for understanding, analyzing and designing neuromimetic networks. It covers the field of statistical learning in pattern recognition, as well as the basics of quantitative assessment of learning. An introduction will be given on deep learning as an extension of multi-layered perceptrons. This course is aimed at students from a variety of disciplines and is designed to be accessible to understanding by students who are not familiar with mathematical techniques.

Contact Marion DOHEN, Anne GUERIN DUGUE

Content(s)

Introduction
Mathematical models: from neuron to networks
Linear models, Associative memories
Nonlinear models, Multilayer Perceptron
Introduction to Deep Learning
Self-organizing networks
Sources separation



Prerequisites

Basic knowledge of mathematics

Test

Written Exam (Exam1, Exam2)
CC: Computer exercice



SESSION 1 (présentiel) : 30% CC + 70% exam
SESSION 2 (présentiel) : 30% CC (session 1) + 70% exam
SESSION 1 (à distance) : 30% CC + 70% exam
SESSION 2 (à distance) : 30% CC (session 1) + 70% exam

Additional Information

Course list
Curriculum->Master->Semester 9
Curriculum->Double-Diploma Engineer/Master->Semester 9

Bibliography

Apprentissage statistique, G. Dreyfus, J.M. Martinez, M. Samuelides, M.B. Gordon, F. Badran, S. Thiria, Eyrolles, 2008