Introduction to Artificial Intelligence
Level Master 1 – Semester S1 - Credits 3 ECTS - Code MU4RBI04
Master Program: Control Science and Robotics
Outline :
This course introduces the basic elements of artificial intelligence (AI). After a general presentation of this discipline, the course is focusing on basic principles, scientific algorithms and practical methodologies useful to engineers to understand, deal and integrate different aspects of AI: knowledge representation, resolution of problems, planning, reasoning on uncertainty and learning. The labs will be using a high-level programming language and a state-of-the-art project in the AI area will be developed (including a coding prototype) and presented at the end of the semester.
Course syllabus :
Courses (8 courses x 2h / course = 16h of courses)
- General introduction to Artificial Intelligence
- Markov decision process, basic learning and reinforcement learning
- Action planning, reasoning on uncertainty, Bayesian networks
- Supervised (methods, bagging, boosting, WMA) and unsupervised (clustering) learning
- Decision tree, random forest, ensemble learning
- Problem solving, state space search (A*, best-first, UCT)
- Stochastic optimization in R^n
- Propositional logic, first-order logic
Labs (4 sessions x 3h = 12h of labs)
- Markov Decision Process (MDP)
- Bayesian network
- Supervised and unsupervised learning (including random forest and bagging))
- Stochastic optimization
Pre-requisites :
Basic mathematics, basic probabilities, basic programming skills
Resources for students:
Course’s slides on Moodle. Lecture notes, Labs notes.
Scientific knowledge developed in the course:
Modern knowledge representation, first-order logic, Markov processes, Bayesian networks, Super-vised / unsupervised learning, ensemble learning, problem solving, stochastic optimisation.
Acquired competencies:
• Deep understanding of the modern knowledge representation
• Understanding the impact of artificial intelligence technologies in the modern world
• Fast prototyping applications related to artificial intelligence
• Understanding the trends of these technologies in the future of engineering (industry x.0)
Evaluation: one final written exam (ER) (50%) and one continuous monitoring (CC) (50%)
Bibliography:
1. Artificial Intelligence : A Modern Approach
Stuart Russell & Peter Norvig, Prentice Hall, Fourth edition, 2018
https://people.engr.tamu.edu/guni/csce421/files/Al_Russell_Norvig.pdf
http://aima.cs.berkeley.edu/
2. Reinforcement Learning : An Introduction
Richard S. Sutton et Andrew G. Barto, MIT Press, second edition, 2020
http://incompleteideas.net/book/RLbook2020trimmed.pdf
Level Master 1 – Semester S1 - Credits 3 ECTS - Code MU4RBI04
Master Program: Control Science and Robotics
Outline :
This course introduces the basic elements of artificial intelligence (AI). After a general presentation of this discipline, the course is focusing on basic principles, scientific algorithms and practical methodologies useful to engineers to understand, deal and integrate different aspects of AI: knowledge representation, resolution of problems, planning, reasoning on uncertainty and learning. The labs will be using a high-level programming language and a state-of-the-art project in the AI area will be developed (including a coding prototype) and presented at the end of the semester.
Course syllabus :
Courses (8 courses x 2h / course = 16h of courses)
- General introduction to Artificial Intelligence
- Markov decision process, basic learning and reinforcement learning
- Action planning, reasoning on uncertainty, Bayesian networks
- Supervised (methods, bagging, boosting, WMA) and unsupervised (clustering) learning
- Decision tree, random forest, ensemble learning
- Problem solving, state space search (A*, best-first, UCT)
- Stochastic optimization in R^n
- Propositional logic, first-order logic
Labs (4 sessions x 3h = 12h of labs)
- Markov Decision Process (MDP)
- Bayesian network
- Supervised and unsupervised learning (including random forest and bagging))
- Stochastic optimization
Pre-requisites :
Basic mathematics, basic probabilities, basic programming skills
Resources for students:
Course’s slides on Moodle. Lecture notes, Labs notes.
Scientific knowledge developed in the course:
Modern knowledge representation, first-order logic, Markov processes, Bayesian networks, Super-vised / unsupervised learning, ensemble learning, problem solving, stochastic optimisation.
Acquired competencies:
• Deep understanding of the modern knowledge representation
• Understanding the impact of artificial intelligence technologies in the modern world
• Fast prototyping applications related to artificial intelligence
• Understanding the trends of these technologies in the future of engineering (industry x.0)
Evaluation: one final written exam (ER) (50%) and one continuous monitoring (CC) (50%)
Bibliography:
1. Artificial Intelligence : A Modern Approach
Stuart Russell & Peter Norvig, Prentice Hall, Fourth edition, 2018
https://people.engr.tamu.edu/guni/csce421/files/Al_Russell_Norvig.pdf
http://aima.cs.berkeley.edu/
2. Reinforcement Learning : An Introduction
Richard S. Sutton et Andrew G. Barto, MIT Press, second edition, 2020
http://incompleteideas.net/book/RLbook2020trimmed.pdf
- Enseignant: Nicolas Bredeche
- Enseignant: Alice Cohen Hadria
- Enseignant: Gabriel Jimenez
- Enseignant: Daniel Racoceanu