GAussian Processes for computer experiments and machine learning: more guarantees and broader applications

Project funding and duration:

The project is funded by the ANR from 2021 to 2025 (€205 000), thanks to the "jeunes chercheurs jeunes chercheuses" instrument (JCJC).

Project members:

• François Bachoc, Institut de Mathématiques de Toulouse (principal investigator)
  
• Ismaël Castillo, Sorbonne Université
  
• Nicolas Durrande, Secondmind, Cambridge
  
• Andrés Felipe Lopez Lopera, Université Polytechnique Hauts-de-France
• Agnès Lagnoux, Institut de Mathématiques de Toulouse
• Pierre Neuvial, Institut de Mathématiques de Toulouse
• Thibault Randrianarisoa, Sorbonne Université
  
• Olivier Roustant, Institut de Mathématiques de Toulouse
• Mark van der Wilk, Imperial College London
  
• Mathis Deronzier, PhD student, Institut de Mathématiques de Toulouse
  
• Elie Odin, PhD student, Institut de Mathématiques de Toulouse
  

Summary of the project:

This project aims for new mathematical guarantees and applications for Gaussian processes. The four main themes are the following.

• Deep Gaussian processes
• Variational inference
• Constrained Gaussian processes
• New applications

How to acknowledge funding from the project:

• This work was supported by the Project GAP (ANR-21-CE40-0007) of the French National Research Agency (ANR).