Elias Ramzi, Nicolas Thome, Clément Rambour, Nicolas Audebert, Xavier Bitot: Robust and decomposable average precision for image retrieval. Advances in Neural Information Processing Systems 34 (NeurIPS, 2021).
- Ph.D in Deep Learning, Le Cnam (France), 2021-2024 (expected)
- Engineering degree, CentraleSupelec (France), 2016-2020
- Preparatory classes, Lycée Lakanal (France), 2014-2016
- 2021-2024: Deep learning PhD student - Le Cnam, Paris - France
- Image retrieval
- Direct optimization of metrics
- Mai to October 2020: Research intern in deep learning - Le Cnam, Paris - France
- Semantic segmentation of medical 3D volumes, using a multi-view approach
- Implementation of a 2D fully convolutional segmentation network: U-Net
- Research on a confidence based fusion with a confidence measure learned by an auxiliary neural network based on ConfidNet
- January to July 2019: Data engineer and data scientist - Sancare, Paris - France
- Data scientist (2 months): Implementing methods for the interpretability of a deep neural network.
- Data engineer (4 months): Developing data pipelines to extract data from hospitals’ IT system and structure them to Sancare’s data model.
- July to December 2018: Data engineer - Balto, Sydney - Australia
Elias Ramzi, Nicolas Audebert, Nicolas Thome, Clément Rambour, Xavier Bitot: RHierarchical Average Precision Training for Pertinent Image Retrieval. In: European Conference on Computer Vision. Springer (ECCV, 2022).
Marc Lafon, Elias Ramzi, Clément Rambour, Nicolas Thome: Hybrid Energy Based Model in the Feature Space for Out-of-Distribution Detection. International Conference on Machine Learning (ICML 2023).
Elias Ramzi, Nicolas Audebert, Clément Rambour, André Araujo, Xavier Bitot, Nicolas Thome: Optimization of Rank Losses for Image Retrieval. Submitted to IEEE Transactions on Pattern Analysis and Machine Intelligence (under-review TPAMI).