Professor of Communication Networks with specialization in stochastic optimization and learning
The field of machine learning aims to build computer systems that are automatically improved by their experiences. In recent decades, machine learning has given us effective web search engines, self-driving cars, technologies for speech and image recognition. It has also helped us to better understand the human genome. But despite these impressive practical advances, a solid mathematical basis is still missing. There is a gap between practice and theory, where the fundamental theoretical knowledge regarding how machine learning works is largely unknown.
Alexandre Proutiere’s research focuses on the development of mathematical tools, in order to create optimised machine learning algorithms. He especially wishes to lay the foundation for dimensional reduction techniques, which can be applied in the analysis of large volumes of data. This includes reporting of the internet users' web browsing activities, in order to map their preferences and recommend products they might be interested in.
The tools that are being developed by Proutiere and his research team can also improve the management of dynamic systems. These can be used for the design of autonomous systems that need to learn from their very complex surroundings, and be able to make decisions based on this. Robots and self-driving vehicles are examples of such autonomous systems.