Descriptif
- Understanding the main machine learning methods and algorithms
- Being able to apply them to computer networks and applications to solve practical use-cases
- Being able to define and follow a correct protocol (data pre-processing, training, test, validation) and to adapt it to the different use-cases
- Being able to use the main Python libraries for Machine Learning
Lecturers: Andrea Araldo (TSP)
30 heures en présentiel
Diplôme(s) concerné(s)
Parcours de rattachement
Format des notes
Numérique sur 20Littérale/grade européenProgramme détaillé
1. Introduction (supervised / unsupervised Machine Learning, protocol, data preparation in python)
2. Data exploration, Linear Regression, evaluation of regression models
3. Neural Networks (application to network intrusion classification)
4. Anomaly detection (application to intrusion detection)
5. Recommender systems (application to recommendation of web content)
6. Time series, Preventive Maintenance, Long Short-Term Memory networks (application to IoT or data centers).
7. Project presentation and exam
All courses will be “cours intégrés”
Evaluation
50% project, 25% exam, 25% participation in class.
