Mastering the basic techniques of Machine Learning and Deep Learning
Pre-requisite skills: Probability / Statistics, Python language.
On completion of the unit, the student will be capable of: | Classification level | Priority |
---|---|---|
Knowing the paradigms of learning | 1. Knowledge | Essential |
Understanding and applying functions of major learning methods, such as Deep Learning | 2. Understand | Important |
Understanding and applying functions of major learning methods, such as Deep Learning | 3. Apply | Important |
Analysing problems such as Data Analysis | 4. Analyse | Important |
Knowing the security flaws of Machine Learning models | 1. Knowledge | Useful |
Percentage ratio of individual assessment | Percentage ratio of group assessment | ||||
---|---|---|---|---|---|
Written exam: | 50 | % | Project submission: | % | |
Individual oral exam: | % | Group presentation: | % | ||
Individual presentation: | % | Group practical exercise: | % | ||
Individual practical exercise: | 50 | % | Group report: | % | |
Individual report: | % | ||||
Other(s): % |
Type of teaching activity | Content, sequencing and organisation |
---|---|
Course | Introduction to Machine Learning, mathematical bases and Python language revision (3h) Learning paradigms Data bases, models, overfitting, generalisation Bases of probability and statistics Revision of Python for data analysis |
Course/Supervised study | Supervised and non-supervised learning (SVM, boosting) (3h) Analysis of principle components Clustering methods Logistic regression SVM Boosting |
Course | (3h) Deep Learning I Perceptron and neural networks MultiLayer Perceptron Convolutionnal Neural Network (CNN) |
Course | (3h) Deep Learning II Recurrent Neural Network (RNN) Autoencoder Generative models Transfer Learning and domain adaptation |
Supervised study | Data analysis / Machine Learning (3) |
Course | Security of Machine Learning (1.5h) Potential threats to the integrity of confidentiality and accessibility to Machine Learning systems Adversarial Examples |
Supervised study | Machine Learning (1.5h) |