The Artificial Intelligence Summer School aims to introduce students to the field of artificial intelligence:

• by covering a wide range of related topics, methodologies and applications;

• by giving students the opportunity to gain practical experience in dealing with real-world problems;by organizing educational visits to companies and institutions where fundamental research in artificial intelligence is taking place.

We hope that by the end of the AI Summer School, students will have :

• a good understanding of key topics and methodologies in machine learning and artificial intelligence in general, including deep learning, computer vision, reinforcement learning, natural language processing and network analysis;

• strengthened their Python programming skills by implementing and evaluating various machine learning algorithms and working on a real-world data challenge.

Python and programming skills are required for the program. If you're not familiar with Python, you'll need to learn the basics before the summer school starts. To prove their Python skills, students will need to provide a completed project demonstrating their programming skills in the registration section.

• Introduction to artificial intelligence
  This course will give an overview of the vast field of artificial intelligence, first tracing its history and then presenting the main approaches to artificial intelligence: reflex-based, state-based, variable-based and logic-based. Students will apply these approaches in a hands-on session.
   

• Introduction to machine learning (Richard Combes)
  This course presents the foundations of statistical learning theory, from fundamental limits to algorithms, their analysis and practical implementation. Important concepts such as Vapnik-Chervonenkis theory, empirical risk minimization, stochastic gradient descent algorithms and Kernel methods will be presented in detail. A laboratory session will enable students to apply the theoretical content to data.
   

• Machine learning on networks (Jhony Giraldo)
  Networks (or graphs) have become ubiquitous, as data from a variety of disciplines can naturally be mapped onto graph structures. Typical examples include social networks (Facebook, Twitter, etc.), information networks (Web, etc.) and technological networks (Internet, etc.). The problem of extracting meaningful information from large-scale graph data in an efficient way has become crucial and challenging with several important applications and, to this end, graph mining and learning methods are leading tools. The aim of this course is to present recent, state-of-the-art methods and algorithms for the analysis, exploration and learning of large-scale network data, as well as their practical applications in various fields (e.g. the web, social networks, recommender systems).
   

• Generative AI and natural language processing (Hakim Benkirane)
   

• Introduction to deep learning (Stergios Christodoulidis)
  This course will provide students with the principles of representation learning and deep learning by covering the following topics: Neural networks, backpropagation and stochastic gradient optimization, auto-encoders, hyperparameters and learning tricks for neural networks, regularization, deep belief networks and deep Boltzmann machines. Students will apply these approaches in a hands-on session.
   

• Introduction to computer vision (Hugues Talbot)
  In this course, students will study the main aspects of the use and techniques of artificial intelligence in the context of medical imaging. Medical imaging is a fascinating and major field of application for artificial intelligence and machine learning, where the stakes are high and the impact on society can be lasting and strong. However, it is also an application area that presents many challenges. In particular, it is important that AI results in this field are reliable, reproducible and verifiable. Doctors are generally unhappy with “black box” decisions that come with no explanation. Fortunately, there are many innovative techniques for producing exactly this type of result, which will be examined in detail.
   

• Introduction to computer vision and medical imaging (Hugues Talbot)
   

• Reinforcement learning (Stergios Christodoulidis)
   

• Artificial intelligence and explicability (Wassila Ouerdane)
   

• Artificial intelligence and the environment (Gilles Faÿ)