The course "Statistical Models for Machine Learning" addresses the problem of machine learning from the perspective of probabilistic models and statistical estimation.
While the course presents useful models and methods in this context, it is not intended to be an exhaustive catalog.
The objective of the course is more to present within a consistent theory the theoretical concepts and tools common to all these models and methods and to show how, based on modeling assumptions specific to each type of problem addressed, these concepts are logically assembled before arriving at an operational learning method.
The challenge is not only to empower students to understand and use existing methods wisely, but also to design new methods (or adapt existing methods) to address the peculiarities of new problems.
The course will also aim to achieve a continuum from theory to practice, whether during lectures or labworks : the hypothesis associated with a given class of problems are first identified, followed by a theoretical modeling work, which leads to the definition of a model and its inference algorithms. These results are then implemented (in Python) and evaluated on data.
- Be able to choose a statistical model/method adapted to the problem under consideration and implement it appropriately
- Be able to understand the theoretical concepts underlying a statistical inference method presented in a scientific article.
- Be able to implement a model / statistical method in a language such as Python.
- Be able to adapt a model/method to take into account the specificities of the problem being addressed.