UE Advanced learning models

Degrees incorporating this pedagocial element :


Statistical learning is about the construction and study of systems that can automatically learn from data. With the emergence of massive datasets commonly encountered today, the need for powerful machine learning is of acute importance. Examples of successful applications include effective web search, anti-spam software, computer vision, robotics, practical speech recognition, and a deeper understanding of the human genome. This course gives an introduction to this exciting field, with a strong focus on kernels methods and neural network models as a versatile tools to represent data

This course deals with:

 Topic 1: Neural networks

- Basic multi-layer networks
- Convolutional networks for image data
- Recurrent networks for sequence data
- Generative neural network models


 Topic 2: Kernel methods

- Theory of RKHS and kernels
- Supervised learning with kernsl
- Unsupervised learning with kernels
- Kernels for structured data
- Kernels for generative models

It is composed of 18 hours lectures.

Evaluation :

There will be a written homework with theoretical exercises. In addition the students participate in a data challenge in which they implement a machine learning method of choice to solve a prediction problem on a given dataset. Both elements contribute equally to the final grade.

See course website.


Fundamental notions in linear algebra and statistics.

Basic programming skills to implement a machine method of choice encountered in the course from scratch