Artificial Intelligence: from Machine Learning to Data Science

Machine Learning addresses the question of how to develop computer systems that improve their performance automatically with experience. It is the core technologies at the basis of modern Artificial Intelligence and of all of its spectacular breakthrough in the last ten years. It is also at the core of the emerging field of Data Science. This course deals with the foundations of machine learning based Artificial Intelligence and of its more recent success which has led to the development of Data Science. The whole process allowing the extraction of knowledge from data will be presented: data preparation and visualization, model generation via the use of a Machine Learning framework, results analysis. The course mainly focuses on classification tasks: non-supervised via the learning algorithm known as k-means, supervised via the presentation of several learning algorithms (k-nearest neighbors, random forest, support vector machines, artificial neural networks). A particular attention is given to practical work: the detailed process (dimension reduction, cross-validation, visualization, fine tuning of hyper-parameters, ...) is studied and well-known toolboxes are introduced (Scikit-Learn, Keras).

The course will be given in English. It is possible to be in French if all the PhD Students registered to the course are francophones