Artificial intelligence

linfo1361  2024-2025  Louvain-la-Neuve

Artificial intelligence
5.00 credits
30.0 h + 30.0 h
Q2
Teacher(s)
Deville Yves; Piette Eric (compensates Deville Yves);
Language
French
Prerequisites
LEPL1402: Programming in a high-level language
Main themes
  • Research-based problem solving: problem formulation, informed and uninformed research strategies, local research, behavioral assessment and estimated cost, applications
  • Constraint satisfaction: formulation problems, constraint tracing and propagation, applications
  • Games and adversarial research: minimax algorithm and Alpha-Beta pruning, applications
  • Propositional logic: knowledge representation, inference and reasoning, applications
  • First-order logic: knowledge representation, inference and reasoning, forward and backward chaining, rule-based systems, applications
  • Planning: planning problem languages, research methods, planning graphs, hierarchical planning, extensions, applications
  • AI, philosophy and ethics: "can machines act intelligently?", "can machines really think?", ethics and the risks of artificial intelligence, the future of artificial intelligence
Learning outcomes

At the end of this learning unit, the student is able to :

With regard to the AA reference of the "Master's degree in computer science" program, this course contributes to the development, acquisition and evaluation of the following learning outcomes:
        INFO1.1-3
        INFO2.2-4
        INFO5.2, INFO5.5
        INFO6.1, INFO6.4
With regard to the AA reference of the "Master [120] in computer science" program, this course contributes to the development, acquisition and evaluation of the following learning outcomes:
        SINF1.M4
        SINF2.2-4
        SINF5.2, SINF5.5
        SINF6.1, SINF6.4
With regard to the AA reference of the "Master [60] in computer science" program, this course contributes to the development, acquisition and evaluation of the following learning outcomes:
        1SINF1.M4
        1SINF2.2-4
        1SINF5.2, 1SINF5.5
        1SINF6.1, 1SINF6.4
Students who successfully complete this course will be able to
  •         explain and make good use of the basic concepts of knowledge representation, problem solving and reasoning methods, as used in artificial intelligence
  •         assess the applicability, strengths, and weaknesses of knowledge representation, problem solving, and reasoning methods in solving real-world engineering problems
  •         develop intelligent systems by assembling solutions to concrete problems
  •         discuss the role of knowledge representation, problem solving and reasoning methods in the design and realization of intelligent systems
Students will have developed methodological and operational skills. In particular, they will have developed their ability to:
  •         master a new programming language primarily using an online tutorial
  •         deal with deadlines and competitiveness when developing an application that wants to be the most efficient.
 
Content
  • Introduction and Intelligent Agents
  • Uninformed and Informed Search
  • Local Search and Heuristics
  • Constraint Satisfaction Problems (CSP)
  • Adversarial Search (Games) and Monte Carlo Tree Search (MCTS) Techniques
  • Logical Agents (First-Order Logic and Inference)
  • Simple Decision Making
  • Complex Decision Making
  • Multi-Agent Decision Making
  • Supervised Learning from Examples
  • Reinforcement Learning
Teaching methods
  • Problem-based learning
  • Learning by doing
  • 3 long-term projects to be completed in pairs (over several weeks)
  • Lectures
  • Theoretical and practical exercises depending on the topic
  • Feedback on completed projects and exercise corrections
Evaluation methods
  • The evaluation will be based on continuous assessment of the missions and assignments completed during the year, as well as a final exam.
  • Continuous assessment includes graded assignments, which will result in a single overall score, communicated at the end of the final assignment. Failure to follow the methodological guidelines defined on Moodle, especially regarding the use of online resources or collaboration between students, will result in a global score of 0 for continuous assessment.
  • The use of ChatGPT or any equivalent tool is strictly prohibited for completing missions and assignments. The professor reserves the right to summon students for an additional oral Q&A session to verify understanding of the submitted work. In case of failure, a global score of 0 will be awarded for the assignment.
  • The weighting of assignments and the final exam is as follows: if all assignments are graded 15/20 or higher, their weight will be 40%, and the exam will count for 60%. If the average of the assignments is below 15/20, their weight will be reduced to 30%, and the exam will count for 70%.
  • Assignments can only be completed during the course term. It is not possible to redo the assignments during another semester or for the August/September session.
  • The exam will be written, but in case of doubt about the grade to be assigned to a student, an additional oral exam may be arranged.
Bibliography
  • Stuart Russell, Peter Norvig, Artificial Intelligence : a Modern Approach, 3nd Edition, 2010, 1132 pages, Prentice Hall
  • transparents en ligne
Faculty or entity


Programmes / formations proposant cette unité d'enseignement (UE)

Title of the programme
Sigle
Credits
Prerequisites
Learning outcomes
Specialization track in Computer Science

Bachelor in Computer Science

Master [120] in Electro-mechanical Engineering

Master [120] in Data Science Engineering

Minor in Computer Sciences

Master [120] in Data Science: Information Technology

Mineure Polytechnique