English

Français

English

Français

Appearance

Artificial Intelligence and Simulation ​

(IAS)

  • Coefficient : 5
  • Hourly Volume: 150.0h (including 72.0h supervised)
    CM : 27h supervised
    Labo : 45h supervised (and 12h unsupervised)
    Out-of-schedule personal work : 66h
  • Including project : 6h supervised and 18h unsupervised project

AATs Lists

Description ​

Introduction to basic artificial intelligence techniques and in-depth study of a chosen theme through a project. Most themes are addressed from an applications perspective and are associated with practical work.

  1. Knowledge Representation
    • Classical Logic Deduction, Logics for Knowledge Representation (modalities, actions and change)
    • Logic Programming
    • Fuzzy Logic and its application to fuzzy control
  2. Data Science - Machine Learning
    • Data Mining
    • Classification
    • Neural Networks
    • Reinforcement Learning
  3. Artificial Intelligence Applications
    • Video Games
    • Autonomous Robotics
    • Natural Language Processing
    • Independent Project Implementation

Learning Outcomes (AAv) ​

  • AAv1 [heures: 10, E1, F1]: By the end of the first module period, students will be able to organize different Artificial Intelligence concepts, methods and techniques and situate and compare them to each other.

  • AAv2 [heures: 20, B1, B2]: By the end of the module, students will be able to name and explain the most appropriate knowledge representation models for formulating and solving problems with varied characteristics.

  • AAv3 [heures: 30, C2, C3]: By the end of the module, students will be able to propose, design and implement a system solving a given problem using a given AI technique.

  • AAv4 [heures: 30, D2, B2]: By the end of the module, students will be able to implement various existing AI-related tools and software libraries for addressed industrial application domains.

  • AAv5 [heures: 20, C4, F1]: By the end of the module, students will be able to analyze and evaluate AI system performance, and identify and take into account potential biases and limitations.

  • AAv6 [heures: 40, C1, C3, D1, F2, G1]: By the end of the module, students will be able to work in teams and independently in designing and implementing a system solving a given problem using appropriate AI techniques of their choice.

Assessment Methods ​

Assessment is done through continuous assessment (at least 3 knowledge tests) and a pair project, with the possibility of retaking one of the continuous assessment grades.

Keywords ​

Logics, knowledge representation and logic programming, fuzzy logic and control, neural networks, reinforcement learning, deep learning, data science, natural language processing (NLP).

Prerequisites ​

Python programming and object-oriented language programming. Courses may be taught in English.

Resources ​

  • Course and tutorial materials (in English)
  • Thematic bibliographies for each theme
  • Stuart Russell and Peter Norvig. Artificial Intelligence. Addison-Wesley, 2010
  • Richard S Sutton and Andrew G Barto, Reinforcement learning: An introduction, 2014.
  • Collective. Artificial Intelligence. What is it really about? Cépaduès Editions, 2020.