University training in Machine Learning

The objective of the training is to offer keys of understanding to people new to the field of machine learning. The training allows to demystify the tools used by data scientists, to manipulate simple models on real data as well as to provide a solid basis for students wishing to pursue further studies in the field.

The course is composed of two learning methods: theoretical sessions with PowerPoint and programming exercises with Python.

This training will aim to introduce the basic concepts of machine learning to a public totally new to the subject. We will have both a theoretical part with slides and a practical part where we will use the python programming language for the exercises. In the theoretical part, we will introduce concepts such as supervised, unsupervised and reinforcement learning.  Some examples of classical algorithms will be presented such as tree-based models, naive Bayesian classification or neural networks. Without trying to formalize too much, the theoretical aspects of machine learning will be presented both intuitively and with the help of a mathematical formalism.  For the programming part, the basic Python packages used in machine learning will be presented, but participants will be assumed to already have a good command of programming (ideally Python). At the end of the course, the objective is to have demystified the tools used by data scientists, to have manipulated simple models on real data, and to have provided a solid foundation for students wishing to pursue their studies in the field.

Daytime