Team : MLIA
Arrival date : 04/03/2017 Localisation : Campus Pierre et Marie CurieSorbonne Université - LIP6 Boîte courrier 169 Couloir 26-00, Étage 5, Bureau 525 4 place Jussieu 75252 PARIS CEDEX 05 FRANCE Tel: +33 1 44 27 51 29, Daniel.Brooks (at) nulllip6.fr
Supervision : Matthieu CORD
Deep learning for radar signals
This PhD aims at developing new deep framework for radar signals. In this context, covariance matrices have attracted attention for machine learning applications due to their capacity to model meaningful statistical properties of the data.
The main challenge is that one needs to take into account the particular geometry of the Riemaniann manifold of symmetric positive definite (SPD) matrices they belong to.
In the context of neural networks, we propose in this PhD to study new deep architectures for handling these SPD matrices. We plan to introduce new projection layers based on the Riemaniann barycenter of the data, or other kinds of projections that could be learned.
D. Brooks, O. Schwander, F. Barbaresco, J.‑Y. Schneider, M. Cord : “Second-order networks in PyTorch”, GSI 2019 - 4th International Conference on Geometric Science of Information, vol. 11712, Lecture Notes in Computer Science, Toulouse, France, pp. 751-758, (Springer) (2019)