Skip to content

Introduction to Machine Learning

PDF icon Download in PDF

  • Lecturer: Catherine ACHARD (catherine.achard@sorbonne-universite.fr)
  • Course code: UM4RBI10-IntroML
  • Student workload: 14h of lectures, h of tutorials, 14h of labs
  • Credits: 3 ECTS
  • Specialization tracks:
  • Semester offered: S1 S2 S3 S4
  • Language of instruction: French English
  • Targeted audience: Eng. Sc. department Other :
  • Localization : PMC Campus Other :

Course Overview

This course introduces the concepts and methods of machine learning. The teaching aims to give students the technical and methodological mastery of the entirety of a learning problem: from the acquisition and preparation of data, the learning of a model, to the evaluation of the models and the interpretation of the results resulting from this evaluation.

Mots-clés : Machine learning. Regression and classification problems. Supervised and unsupervised learning. ACP. KPPV. Random forest. Bayesian classification. SVM. Neural networks. Data preparation. Evaluation methodology.

Prerequisites

Students should have previously acquired the following prerequisites to follow this course:

  • Mathematics: linear algebra, vector and matrix analysis, functions of several variables
  • Computer science: programming (in Python: numpy, etc.), algorithms

Intended Learning Outcomes

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

  1. Analyze a problem: Identify the right paradigm for the task and the data
  2. Analyze a problem: Choose an algorithm that is appropriate for the task, the data, and the resources
  3. Design: Sizing a Model
  4. Design: Specify meta parameters to optimize the training of a model's parameters
  5. Implementation: Use advanced deep learning design environments (language, api, ide, hardware) to implement a model implementation
  6. Implementation: Preparing data for training
  7. Evaluation: Develop an adequate experimental protocol to evaluate the performance of a model
  8. Evaluation: Monitor and interpret the results obtained to ensure optimal learning

Indicative Teaching Sequence and Methods

Week C/TD/TP* Content Preparation Learn.\ outc.
S1 C1 C1 - Intro, coding, covariance matrix, ACP, LDA, feature selection AAV 1
S2 C2 / TP1 C2 - Formalism of a learning problem, performance of a classifier, KPPV and Mahalanobis distance TP1 - ACP AAV 1, 2
S3 C3 / TP2 C3 - Random Forest, Bayesian Classification, Estimation of DDPs TP2 - KPPV with test on 2 unformatted images AAV 1, 2, 5, 6, 7
S4 C4 / TP3 C4 - SVM TP3 - ROC Curve AAV 1, 2, 5, 6, 7
S5 C5 / TP4 C5 - Regression, Unsupervised Classification TP4 - Random forest AAV 1, 2, 3, 4, 8
S6 C6 / TP5 C6 - Introduction to Neural Networks: From MLP to CNN TP5 - Error bias/variance, performance measurement on pre-trained models AAV 1, 2, 3, 4, 5
S7 C7 / TP6 C7 - Classical neural architectures, attention processes, application to U-Net segmentation TP6 - SVM AAV 1, 2, 3, 4, 5
S8 TP7 TP7 - Neural Networks AAV 1, 2, 3, 4, 5, 6, 7
  • C/TD/TP respectively corresponds to lectures, tutorials and lab sessions.

Indicative Assessment of Intended Learning Outcomes (1st session)

Week Individ./group In-person/remote Type of exam Evaluated outcomes Scale %
S4 Individual In-person Written AAV 1- 8 30%
S10 Individual In-person Written AAV 1- 8 30%
S10 Individual In-person Practical AAV 1- 8 20%
S10 Individual In-person Practical AAV 1- 8 20%

2nde session

Session Individ./group In-person/remote Type of exam Evaluated outcomes Scale %
2 Individual In-person Written AAV 1-8 60%
1 Individual In-person Practical AAV 1-8 40%

Logo SDI Date of generation of this unit description: 14/01/2026 Logo SDI