The rapid growth of Artificial Intelligence (AI) research and applications offers unprecedented opportunities. This course is intended for students wishing to receive an excellent primary education covering a broad spectrum of concepts and applications of data-driven AI and learning from examples.
The program offers introductory courses in statistical learning, deep learning and reinforcement learning, optimization, signal processing, information theory, and game theory. Numerous options make it possible to perfect oneself in learning theory and to specialize in many fields such as big data, image, and language processing.
This second-year offers an expanded choice of options, covering ethical aspects and other topics such as starting a company.
This course requires a good background in mathematics and computer science: - Probability and statistics - Linear algebra - Differential and integral calculus - Scientific programming - Visualization of the data Applicants should also have completed the M1 of Artificial Intelligence (or equivalent) successfully: - Know the basics of applied statistics and optimization - Know how to manipulate big data - Know how to differentiate and apply techniques of supervised, unsupervised, and reinforcement learning - Know how to program predictive models with Python and master sci-kit-learn - Know how to visualize data and illustrate results with programming tools - Know how to write a project proposal and communicate results in writing and orally.
Skills:
Mathematically formulate gradient descent algorithms for deep neural networks, graphical models, or other statistical learning models.
Program deep learning models and graphical models using Python and acquire proficiency in Keras, TensorFlow, and Pytorch.
Understand the foundations of statistical learning at a theoretical level, focussing on over-learning and regularisation.
Analyze data of various types (image, text, speech) from the raw signal.
Read, summarise, comment on and reproduce scientific articles.
Career prospects:
This course prepares to research and R & D professions in new fields of application in full swing: computer vision (autonomous vehicles and biometrics); voice recognition (necessary for new human-machine interfaces for smartphones); filtering and aggregation of heterogeneous and textual content (essential to commercial solutions for managing significant data streams); management and monitoring of complex or critical industrial systems that rely on data analysis.