Chair objectives
Representation learning is a fundamental aspect of artificial intelligence that involves automatically extracting a compact and useful representation of input data for classification, clustering, and prediction tasks.
However, this can be a challenging task in deep learning, where overparameterization is a concern, and in high-dimensional statistics, where sparse structures may be hidden.
In many industrial settings, computer simulations generate complex data that require feature selection for effective analysis.
To overcome this challenge, we propose leveraging powerful mathematical tools for efficient representation selection, including Sensitivity Analysis (SA), signatures from Rough Path Theory (RPT), and Random Matrix Theory (RMT).
Sensitivity analysis enables the identification of the most informative features by selecting the variables that contribute the most to the variability in the data.
Signatures from Rough Path Theory provide a universal and expressive representation for time series data that captures complex temporal dependencies and enables
effective analysis.
Finally, RMT can estimate the number of significant principal components in highdimensional datasets and provide insight on the stability of neural networks.
By utilizing these mathematical tools for representation learning, our goal is to facilitate efficient and effective analysis of complex industrial data, ultimately supporting data-driven decision making
- Sensitivity analysis for variable selection and understanding black-box computer codes
- Random Matrix Theory for large dimensional statistics, more precisely for signal recovery and
stability of neural networks - Rough path theory for universal representations of time series
- NXP SEMICONDUCTORS High dimensional sensitivity analysis and Gaussian processes. Cifre thesis of D. REMOT (Sept 2022-...), supervised by R. CHHAIBI and C. PELLEGRINI .
- SAFRAN TECH Generation of 3D shapes for industrial blade design. Cifre thesis of V. FOY (Sept 2020-...), supervised by R. CHHAIBI and F. GAMBOA.
- RENAULT Machine learning for road conditions. Cifre thesis of V. NGUYEN (2022-...), supervised by R. CHHAIBI, F. GAMBOA, S. GRATTON, S. ZHANG (Sept 2022-...).
- EDF Reliability and information geometry. Cifre thesis of B. KETEMA (2022-...), supervised by F. GAMBOA and F. Costantino.
- LIEBHERR Machine learning and virtual sensing. Cifre thesis of J. REVERDI (2021-...), supervised by F. GAMBOA and S. Gratton
- EDF Game theory and sensitivity analysis. Cifre thesis of M. IL IDRISSI (2021-...), supervised by F. GAMBOA and J.M. Loubes
- More industrial projects are currently in discussion with AIRBUS, RENAULT and the CEA..
- Learning compact and expressive representations
- Enhancing classical IA methods with novel mathematical tools
English 
French