Aller à : contenu haut bas recherche
 
 
EN     FR
Vous êtes ici:   UNIL > HEC Inst. > HEC App. > SYLLABUS
 
 

Deep Learning

  • Enseignant(s):   I.Rudnytskyi  
  • Titre en français: Apprentissage approfondi
  • Cours donné en: anglais
  • Crédits ECTS: 3 crédits
  • Horaire: Semestre de printemps 2019-2020, 2.0h. de cours (moyenne hebdomadaire)
  •  séances
  • site web du cours site web du cours
  • Formations concernées:
    Maîtrise universitaire ès Sciences en management, Orientation marketing

    Maîtrise universitaire ès Sciences en management, Orientation business analytics

    Maîtrise universitaire ès Sciences en management, Orientation stratégie, organisation et leadership

    Maîtrise universitaire ès Sciences en management, Orientation comportement, économie et évolution

 

Objectifs

By the end of this course, students must be able to:

  • setup a local/remote workstation for working with artificial neural networks (ANN) using R, {keras} package, and backend library TensorFlow
  • use R package {keras} to manipulate ANN models: build, train, tune hyperparameters, save, and use pretrained neural networks
  • apply ANN in the context of image recognition and text mining to the cases covered during the course
  • discuss, evaluate, and visualize results provided by ANN models

Contenus

  1. Foundations of ANN: linear regression, perceptron, gradient descent, and backpropagation
  2. {keras} and TensorFlow: implementation and manipulation of ANN models
  3. Training deep learning networks: hyperparameters tuning, optimization algorithms, activation functions
  4. ANN with dense layers and its applications
  5. Convolutional Neural Networks and their applications in image recognition and computer vision
  6. Recurrent Neural Networks and their applications in text mining and time series

Références

  • Chollet, F. and Allaire, J.J. (2018). Deep Learning with R. Manning Publications Co.
  • Géron, A. (2019). Hands-On Machine Learning with Scikit-Learn, Keras & TensorFlow: Concepts, Tools, and Techniques to Build Intelligent Systems, Second Edition. O'Reilly Media, Inc.

Pré-requis

Students of Master in Management (Business Analytics orientation):

  • Machine Learning In Business Analytics

External students:

  • foundations of supervised machine learning (e.g., data splitting strategies, metrics, and scoring, etc.)
  • basic calculus knowledge (e.g., derivatives, gradient, etc.)
  • essential skills in R

Evaluation

1ère tentative

Examen:
Sans examen (cf. modalités)  
Evaluation:

The assessment is based on two parts, namely homework (50%) and a course project (50%). The maximum number of points that can be obtained in this course is 60 points. Each of the two homework is worth 15 points, while the remaining 30 points are allocated to the course project.

The total number of points is then rescaled to the final grade using the scale below:

6.0 | 57-60

5.5 | 52-56

5.0 | 47-51

4.5 | 42-46

4.0 | 37-41

3.5 | 32-36

3.0 | 27-31

2.5 | 22-26

2.0 | 17-21

1.5 | 12-16

1.0 | 0-11

Rattrapage

Examen:
Sans examen (cf. modalités)  
Evaluation:

A second attempt is solely based on the project (60 points, the same rescaling system is used as in the first attempt). Students should implement proposed modifications to the project (i.e., the report) and present the project again.



[» page précédente]           [» liste des cours]
 
Recherche


Internef - CH-1015 Lausanne - Suisse  -   Tél. +41 21 692 33 00  -   Fax +41 21 692 33 05
Swiss University