Expanding boundaries of Gap Safe screening
Sparse optimization problems are ubiquitous in many fields such as statistics, signal/image
processing and machine learning. This has led to the birth of many iterative algorithms to …
processing and machine learning. This has led to the birth of many iterative algorithms to …
Path following algorithms for -regularized -estimation with approximation guarantee
Y Zhu, R Liu - Advances in Neural Information Processing …, 2023 - proceedings.neurips.cc
Many modern machine learning algorithms are formulated as regularized M-estimation
problems, in which a regularization (tuning) parameter controls a trade-off between model fit …
problems, in which a regularization (tuning) parameter controls a trade-off between model fit …
Enhancing safe screening rules with adaptive thresholding for non-overlapping group sparse norm regularized problems
H Chahuara, P Rodriguez - 2023 24th International Conference …, 2023 - ieeexplore.ieee.org
Sparsity is an often desired property in machine learning and signal processing problems.
Recently, techniques such as screening rules were proposed to exploit sparsity in order to …
Recently, techniques such as screening rules were proposed to exploit sparsity in order to …
A generic coordinate descent solver for non-smooth convex optimisation
O Fercoq - Optimization Methods and Software, 2021 - Taylor & Francis
We present a generic coordinate descent solver for the minimisation of a non-smooth
convex objective with structure. The method can deal in particular with problems with linear …
convex objective with structure. The method can deal in particular with problems with linear …
Quelques contributions dans la conception de “régions sûres” et “tests d'élagages sûrs” en optimisation convexe
T Le Tran - 2023 - theses.hal.science
Convex optimization is common in machine learning, statistics, signal, and image
processing. Solving high-dimensional optimization problems remains challenging due to …
processing. Solving high-dimensional optimization problems remains challenging due to …
A greedy screening test strategy to accelerate solving LASSO problems with small regularization parameters
In the era of big data remarked by high dimensionality and large sample size, the least
absolute shrinkage and selection operator (LASSO) problems demand efficient algorithms …
absolute shrinkage and selection operator (LASSO) problems demand efficient algorithms …