Ayoub Belhadji
Postdoctoral Researcher
abelhadj(at)mit.edu
Room 45-421F
MIT, Buiding 45
51 Vassar St.
Cambridge, MA 02139
Home |
Software
About me
I am currently a Postdoctoral Research Associate at Uncertainty Quantification Group at MIT, working with Youssef Marzouk.
I am interested in problems that lie at the intersection of applied mathematics, signal processing, and machine learning. Currently, I am researching new representations of sets and probability measures, with a focus on their application in the field of optimal experimental design. I completed my Ph.D. in theoretical signal processing and machine learning at Centrale Lille, under the supervision of Pierre Chainais and Rémi Bardenet. My work includes subset selection problems, quadrature rules and interpolation methods using determinantal point processes. Before joining MIT, I collaborated with Rémi Gribonval on compressive learning using sketching methods at ENS de Lyon.
Preprints
Nonlinear Bayesian optimal experimental design using logarithmic Sobolev inequalities
Fengyi Li, Ayoub Belhadji, Youssef Marzouk.
Signal reconstruction using determinantal sampling
Ayoub Belhadji, Rémi Bardenet, Pierre Chainais.
Journals
Sketch and shift: a robust decoder for compressive clustering
Ayoub Belhadji, Rémi Gribonval.
To appear in Transactions on Machine Learning Research
Revisiting RIP guarantees for sketching operators on mixture models
Ayoub Belhadji, Rémi Gribonval.
Journal of Machine Learning Research 25 (55), 1-68.
A determinantal point process for column subset selection
Ayoub Belhadji, Rémi Bardenet, Pierre Chainais.
Journal of Machine Learning Research 21 (197), 1-62.
Conferences
An analysis of Ermakov-Zolotukhin quadrature using kernels
Ayoub Belhadji.
NeurIPS, 2021, Online.
Kernel interpolation with continuous volume sampling
Ayoub Belhadji, Rémi Bardenet, Pierre Chainais.
International Conference on Machine Learning, 2020, Online.
Kernel quadrature with DPPs
Ayoub Belhadji, Rémi Bardenet, Pierre Chainais.
NeurIPS, 2019, Vancouver, Canada.
Des règles de quadrature dans les RKHSs à base de DPPs
Ayoub Belhadji.
XXVIIIème Colloque francophone de traitement du signal et des images
(GRETSI)
September 2022, Nancy (France)
Un processus ponctuel déterminantal pour la sélection d'attributs
Ayoub Belhadji, Rémi Bardenet, Pierre Chainais.
GRETSI, 2019, Lille, France.
Manuscript
Subspace sampling using determinantal point processes
Ayoub Belhadji.
Ph.D. thesis, Centrale Lille Institut, 2020