Minor Introduction to Machine Learning

Coordinator

Michel Riveill

SEMESTER

  • Autumn 2020
  • Spring 2021

FORMAT

Campus SophiaTech, Sophia Antipolis

Prerequisites

Basic programming if possible with Python

Lecturers
 
Autumn 2020
Spring 2021

Learning Outcomes

At the end of the course, students should be able to deal with basic problems in particular in the context of regression, classification, clustering or dimensionality reduction. They must also be able to explain the main differences between these different algorithms.
In the laboratories we will use the Python language and the sklearn library.

This minor gives you the keys to understanding the issues in the field and the tools to deal with simple data sets.​
  • It emphasizes how an algorithm works and especially its use​ (it is not on the programmation of the algorithm​).
  • We will place ourselves from the point of view of a user​.
Summary
Autumn 2020 syllabus

Machine learning and data analysis are increasingly at the centre of many sciences and applications. In this course, the fundamental principles and methods of machine learning will be introduced, analysed and put into practice.                                                                                
The main topics will be presented: linear and logistic regression, the principle of neural network functioning and the multilayer perceptron, unsupervised learning and some elements for working with text in machine learning.  

The content of the autumn semester is more specifically intended for computer science students who are familiar with programming in general and who are able to quickly learn the necessary Python basics.                                                               

Winter school syllabus
Machine learning and data analysis are increasingly at the centre of many sciences and applications. In this course, the fundamental principles and methods of machine learning will be introduced, analysed and put into practice. The main topics will be presented: linear and logistic regression, the principle of neural network functioning and the multilayer perceptron, unsupervised learning and some elements for working with text in machine learning.  
 
The winter semester content is based on the main principles of the autumn semester but is accessible to students with little programming knowledge. The part concerning the preparation of data will not be discussed. Indeed, it is often the part that requires a significant investment in programming. If necessary, during the practical sessions, the code will be given.

Prerequisites:

Even if advanced knowledge of Python is not necessary to attend the Winter School: Introduction to Machine Learning you still need to know the basics of this programming language.

All the exercises will be given with the help of the notebook skirt. The most efficient way is to install it on your anaconda station (https://www.anaconda.com/products/individual). The proposed version installs everything necessary for the course.

Then if you do not practice Python on a daily basis,

Capacity
  • 24 students
Bibliography
  • Statistics and Machine Learning in Python. Edouard Duchesnay, Tommy Löfstedt ​(presents Python language for machine learning​)
  • Python Machine Learning. Sebastian Raschka​
  • Hands-on machine learning with Scikit-Learn and TensorFlow: concepts, tools, and techniques to build intelligent. A. Géron. O’Reilly​ (practical)
  • Data science : fondamentaux et études de cas: Machine learning avec Python et R. M. Lutz, E. Biernat. Editions Eyrolles​ (In French, introduction level but requires some maths background)
Schedule
ATTENTION! The final Autumn 2020 session (Dec,17) has been updated. See below.
 
Autumn 2020

Date

Time slot

Room

Lecturers

Course' title

15 Oct

09:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Rodrigo Cabral Farias

Introduction to Machine Learning

22 Oct

09:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Rodrigo Cabral Farias

Linear Regression​

05 Nov

09:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Rodrigo Cabral Farias

Logistic Regression and first evaluation​

12 Nov

09:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Lionel Fillatre

Decision Tree and Random Forest​

19 Nov

09:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Lionel Fillatre

Introduction to Deep Learning​

26 Nov

08:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Michel Riveill

Unsupervised learning (clustering & dimension reduction)​

3 Dec

08:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Michel Riveill

Text feature extraction​
10 Dec 08:00 - 12:00

SophiaTech
O+315 (Templiers 1)​

Michel Riveill Introduction to recommender systems​
[UPDATE]
17 Dec
08:00 - 12:00 SophiaTech
O+315 (Templiers 1)​
Michel Riveill Final project
Spring 2021

Date

Time slot

Room

Lecturers

Course' title

Jan 11

09h00-16h00

SophiaTech
O+315 (Templiers 1)

Rodrigo Cabral Farias

Introduction, Linear and Logistic Regression

Jan 12

09h00-12h00

SophiaTech
O+315 (Templiers 1)

Rodrigo Cabral Farias

Decision Tree / Random Forest / Naive Bayes / KNN
13h00-16h00 SophiaTech
O+315 (Templiers 1)
Michel Riveill Decision Tree / Random Forest / Naive Bayes / KNN

Jan 13

09h00-16h00

SophiaTech
O+315 (Templiers 1)

Michel Riveill

Clustering, Text processing

Jan 14

09h00-12h00

SophiaTech
O+315 (Templiers 1

Michel Riveill

Recommender Systems
13h00-16h00

SophiaTech
O+110 (Templiers 1

Michel Riveill Project

Jan 15

09h00-12h00

SophiaTech
O+315 (Templiers 1

Michel Riveill

Project (evaluated at the end of the afternoon)

 
Evaluation
Autumn 2020
  • Nov, 5th: Written exam
  • Dec, 17th: Final exam
Spring 2021
  • Jan, 15th: QCM
  • Jan, 31st: Project