I am currently working as a data scientist at Sancare and as a post-doc at the Center for Computational Biology (CBIO) at Mines ParisTech under the supervision of Chloé Azencott, in several projects involving classification of diseases from electronic health records (EHRs), automated patient pre-screening for clinical trials, and privacy-preserving training of distributed models.

Previously, I have worked on solving permutation problems (akin to solving a huge jigsaw puzzle) through relaxation based approaches, applied to de novo genome sequencing, during my Ph.D. in the Sierra Team of the Computer Science Department of École Normale Supérieure, under the supervision of Alexandre d'Aspremont. I graduated from École Normale Supérieure de Cachan (a.k.a., ENS Paris-Saclay) in 2015, where I obtained a Masters Degree in Physics in 2014, and in Mathematics, Machine Learning and Computer Vision (MVA) in 2015. I completed my PhD in October, 2018.

Publications

Relaxations of the Seriation problem and applications to de novo genome assembly
Antoine Recanati
PhD Manuscript
A spectral algorithm for fast de novo layout of uncorrected long nanopore reads
Antoine Recanati, Thomas Brüls, Alexandre d'Aspremont.
In Bioinformatics, 2017, DOI : 10.1093/bioinformatics/btx370
Robust Seriation and Applications to Cancer Genomics
Antoine Recanati, Nicolas Servant, Jean-Philippe Vert, Alexandre d'Aspremont.
Reconstructing Latent Orderings by Spectral Clustering
Antoine Recanati, Thomas Kerdreux, Alexandre d'Aspremont.

Teaching

2016-2017 : Teaching Assistant, Introduction to Programming (C++), École des Ponts ParisTech (Engineering school).
2017-2018 : Substitute Teaching Assistant for a few lessons, same course.

Talks

October 2016, Institut Curie. [slides]
June 2018, Symbiose Seminar, INRIA Rennes. [slides]
October 2018, PhD Defense, INRIA Paris, [slides]
March 2019, CBIO meeting, Mines ParisTech, [slides]