Maria-Luiza Vladarean
Fields of expertise
My fields of expertise are convex, non-convex and stochastic first-order methods. I'm also interested in the properties of ODE discretizations (specifically the pair GF/GD) and the connections between dynamical systems and optimization algorithms.
Biography
I graduated with a Masters in Computer Science and a minor in Biocomputing from EPFL in 2016. I subsequently spent two years in the industry working as a Software Engineer at Amazon Video in London. At the end of 2018 I returned to EPFL to start my PhD in mathematical optimization, for which I've been lucky to receive the guidance of Prof. Nicolas Flammarion.Publications
Selected publications
| Aditya Varre, Maria-Luiza Vladarean, Loucas Pillaud-Vivien, Nicolas Flammarion UNDER REVIEW (NeurIPS 2023) |
On the spectral bias of two-layer linear networks |
| Maria-Luiza Vladarean, Nikita Doikov, Martin Jaggi, Nicolas Flammarion COLT, 2023 |
Linearization Algorithms for Fully Composite Optimization |
| Gideon Dresdner, Maria-Luiza Vladarean, Gunnar Rätsch, Francesco Locatello, Volkan Cevher, Alp Yurtsever AISTATS, 2022 |
Faster One-Sample Stochastic Conditional Gradient Method for Composite Convex Minimization |
| Maria-Luiza Vladarean, Yura Malitsky, Volkan Cevher NeurIPS, 2021 |
A first-order primal-dual method with adaptivity to local smoothness |
| Maria-Luiza Vladarean, Ahmet Alacaoglu, Ya-Ping Hsieh, Volkan Cevher ICML, 2020 |
Conditional gradient methods for stochastically constrained convex minimization |
| *Jérémie Rappaz, *Maria-Luiza Vladarean, Julian McAuley, Michele Catasta WSDM, 2017 |
Bartering Books to Beers: A Recommender System for Exchange Platforms |