MSc in Artificial Intelligence

Program outline

1st period : Foundations with 6 Core Courses

 

  • Foundations of Machine Learning:An overview of the most important trends in machine learning, with a particular focus on statistical risk and its minimization with respect to a prediction function is given in this course. A substantial lab section involves group projects on data science competitions and gives students the ability to apply the course theory to real-world problems.

  • Foundations of Artificial intelligence: An history and overview of the different approaches of Artificial Intelligence: from reflex agent (low level AI) to expert systems and xIA (high level AI). Each notion will be the subject of individual practical work. In addition, an AI will be developed by group and will compete in a tournament.

  • Foundations of Decision modeling: Preferences are present and pervasive in many situations involving human interaction and decisions. Preferences are expressed explicitly or implicitly in numerous applications and relevant decision should be made based on these preferences. This course aims at introducing preference models for multicriteria decisions. We will present concepts and methods for preference modelling and multicriteria decision making.

  • Foundations of Optimization:  Foundational theory and methods for the solution of optimization problems; iterative techniques for unconstrained minimization; linear and nonlinear programming as well as discrete methods for engineering applications associated with Programming exercises in Python are covered in this course.

  • Foundations of Deep Learning: This course will introduce the modern theory of convolutional neural networks, both in terms of theoretical concepts as well as in terms of practice with different training and programming architectures. Concrete examples on various applications domains will demonstrate the interest of these methods in artificial intelligence.

  • Foundations of Big Data & AI Programming Languages & Platforms :  This course will teach you all about big data management - algorithms, techniques and tools needed to support big data processing with emphasis on the computational aspects related with programming of artificial intelligence methods based on machine learning.

 

Our Pr Vincent Mousseau introduces our MSc in #ArtificialIntelligence pic.twitter.com/SpXaomoI1L

— CentraleSupélec (@centralesupelec) 11 avril 2018

 

Theoretical AI: At least 3 electives to choose

 

  • Reinforcement learning: This course will introduce the foundations of dynamical problem modeling in artificial intelligence through reinforcement learning strategies. In particular we will discuss optimization strategies, sampling strategies and rewards selection strategies at the concept and application level for various problems of artificial intelligence.

  • Excellence in Game Theory: This course will  initially present the main principles concerning decision under uncertainty, and the use of graphical models when making decision under uncertainty Second, we will consider principles of game theory and show how such theory can model and analyse decision in situation where uncertain and strategic interactions are involved.

  • Inference and learning of Graphical Models: This course addresses mathematical foundations and computational solutions for training and optimizing (higher order) probabilistic graphical modes. These are powerful middle-level representations that once endowed with efficient optimization algorithms produce state of the art results for problems with average volume of training data.

  • Multi-agent Systems :  The aim of this course is to study multi-agent systems, i.e. systems composed of multiple interacting computing elements, known as agents, as a paradigm for implementing autonomous and complex intelligent systems.

  • Advanced Statistics :  This course aims first at introducing the general methodology of mathematical statistics through the fundamental concepts (statistical modelling and sampling, estimation problems, decision theory and hypothesis testing). Then, this course provides advanced statistical techniques for multivariate analysis with a particular focus on computational statistics and robust estimation approaches. Regularized / penalized techniques are also presented.

  • Advanced Deep Learning : Deep learning methods are now the state of the art in many machine learning tasks, leading to impressive results. Nevertheless, they are still poorly understood, neural networks are still difficult to train, and the results are black-boxes missing explanations. Given the societal impact of machine learning techniques today (used as assistance in medicine, hiring process, bank loans...), it is crucial to make their decisions explainable or to offer guarantees. Besides, real world problems usually do not fit the standard assumptions or frameworks of the most famous academic work (data quantity and quality, expert knowledge availability...). This course aims at providing insights and tools to address these practical aspects, based on mathematical concepts.

 

Applied AI  : At least 3 electives to choose

  • Visual computing: This course will present an overview of trends, modern methods and applications of computer vision technologies in various problems of visual computing, namely visual analytics, object recognition, 3D scene modeling from multiple-views, cross training of multimodal data, etc.

  • Natural language processing: This course addresses fundamental questions at the intersection of human languages and computer science. In this course we explore methods inspired from symbolic and sub-symbolic artificial intelligence towards language understanding, parsing, translation & generation.

  • Networks science analytics: The problem of extracting meaningful information from large scale graph data in an efficient and effective way has become crucial and challenging with several important applications in AI. The goal of this course is to present recent and state-of-the-art methods and algorithms for analyzing, mining and learning large-scale graph data, as well as their practical applications in various domains.

  • Information retrieval and extraction:  This course addresses the fundamentals of Information retrieval, the process of answering to an information need, expressed by an user’s query, by retrieving the relevant information in non-structured data collections, often massive. This course will also cover recent approaches such that semantic web and question answering with knowledge graphs. A substantial practical section involves group projects on the design and building of a search application.

  • Medical Imaging: This course will present an overview of trends, relevant to the automatic interpretation of medical imaging from computer aided solutions. The course will discuss the entire chain of problems in mid and high-level interpretation addressing the pillar problems of the field (detection, segmentation, registration) and the most ai-driven advanced technologies for computer aided diagnosis.

 

3rd period: Internship & Report (4 to 6 months)