Taming Nonconvexity in Structured Low-Rank Optimization
From 01-01-2022 to 31-12-2025
Description
Recent advances have made possible the acquisition, storing, and processing of very large amounts of data, strongly impacting many branches of science and engineering. A way to interpret these data and explore their features is to use techniques that represent them as certain low-dimensional objects that allow for an intuitive interpretation by domain-specific experts. Such techniques typically factorize the data as two or more structured objects—e.g., orthogonal matrices, sparse tensors—with a lower rank than the original data. The factorizations can usually be formulated as solutions to largescale nonconvex optimization problems; it is of interest to develop fast algorithms to solve them and, in particular, algorithms for which one can prove that they always converge to useful solutions. One cannot enforce this guarantee in many methods; however, we argue that recent developments made by the applicants on Bregman proximal envelopes and on block and multi-block relative smooth functions are excellent tools to develop such algorithms. In short, this project aims (i) to introduce and study a very general formulation for nonsmooth, structured low-rank optimization, (ii) to establish conditions under which this formulation is tractable (even if nonconvex), (iii) to design provably convergent algorithms to address it, and (iv) to apply and test the new model and algorithms in problems from two domains: image processing and genomic analysis.
Team
- Masoud Ahookhosh, Co-promoter
- Panagiotis Patrinos, Coordinator
Financing
Funding: FWO - Research Foundation - Flanders
Program/Grant Type: FWO Project - FWO Research Project
Events
2/09/2024:
PhD defense - Martijn Oldenhof
Machine Learning for Advanced Chemical Analysis and Structure Recognition in Drug Discovery
3/09/2024:
Meet the Jury Igor Tetko on Advanced Machine Learning in Drug Discovery
12/09/2024:
Multimodal analysis of cell-free DNA for sensitive cancer detection in low-coverage and low-sample settings
Seminar by Antoine Passemiers
News
STADIUS Alumni Herman Verrelst – new CEO of Biocartis
08 June 2017
Herman Verrelst, the founder of KU Leuven spin-off Cartagenia, who has been working in Silicon Valley, US for the last few years will be returning to Belgium to follow the steps of Rudi Pauwels as CEO of the Belgian diagnostic company, Biocartis.
Supporting healthcare policymaking via machine learning – batteries included!
29 May 2017
STADIUS takes the lead in the data analytics efforts in an ambitious European Project MIDAS.
Marc Claesen gives an interview about his PhD for the magazine of the Faculty of Engineering Sciences "Geniaal"
10 February 2017
Did you know that in Belgium approximately one third of type 2 diabetes patients are unaware of their condition?
Joos Vandewalle is nieuwe voorzitter KVAB
09 October 2016
Op 5 oktober 2016 heeft de Algemene Vergadering van de Academie KVAB Joos Vandewalle verkozen tot voorzitter van de KVAB.