arrow drop search cross
ERROR: You are not allowed to access 'masters' in this context

Introduction to Machine Learning

Minor


With UCA Department of Computer Science

Summary

This minor presents the main machine learning algorithms in a succinct way. It is intended to be followed with a minimum of prerequisites. The goal is to allow, in simple cases, to implement a data analysis workflow and, in more complex cases, to be able to discuss with data scientists and to understand their vocabulary. In addition to the presentation of the main algorithms, we will insist on the preparation of the data and the different ways of representing the same element as well as on the different metrics allowing to evaluate an algorithm.

Main concepts discussed:

The course will introduce the concepts of linear regression and logistic regression. He will present the techniques of data representation (numerical, classes, words), the main algorithms of supervised learning and the clustering algorithms. It will address the neural networks of text analysis and the principle of recommendation systems.

Course content:

Week 1:

Introduction: what is machine learning and what are the affordable problems?

Labs on the use of Jupiter notebooks

Week 2:

Linear regression and logistic regression

Weeks 3 to 6:

Supervised algorithms

Labs implementation of each algorithm studied on a dataset

Weeks 7 and 8:

Clustering algorithms

Labs: cluster when target classes are known and when they are not

Week 9:

Representation of words and text in machine learning workflows

Labs on building a neural network to translate a simple text from French to English

Week 10:

Principle of recommendation systems

Selected references

  • Linear regression and logistic regression :

- https://jalammar.github.io/visual-interactive-guide-basics-neural-networks/

- https://towardsdatascience.com/5-reasons-logistic-regression-should-be-the-first-thing-you-learn-when-become-a-data-scientist-fcaae46605c4

  • Neural networks :

- https://www.college-de-france.fr/site/yann-lecun/inaugural-lecture-2016-02-04-18h00.htm

- https://www.college-de-france.fr/site/yann-lecun/course-2015-2016.htm

  • Deep learning :

- https://www.college-de-france.fr/site/stephane-mallat/inaugural-lecture-2017-2018.htm

- https://www.college-de-france.fr/site/stephane-mallat/course-2017-2018.htm

  • Learning with large dimensions :

- Imagerie médicale et apprentissage automatique : vers une intelligence artificielle ? (https://www.college-de-france.fr/site/gerard-berry/symposium-2017-2018.htm)

 

Prerequisites

  • Use of Jupiter notebooks.
  • Basic programming in Python. A self-assessment notebook will be made available (if you cannot do the exercises presented, basic Python training is required).