Research

My research lies at the interface of Riemannian geometry, statistical learning, and signal processing. It is organised around three connected themes.

• Geometry-aware federated learning Federated training of SPDNet on covariance-based representations of EEG signals, with Riemannian aggregation schemes and privacy guarantees. Current, since 2025.
• Riemannian stochastic optimization Stochastic optimization on Riemannian manifolds. Collaboration with F. Portier (ENSAI)
• Zeros of random trigonometric functions Universality of the expected number of real zeros under dependent and non-Gaussian coefficients. 2019–2022. With G. Poly and J. Angst.