3 papers
math.NA2025
Structural Packing and Dyadic Factorization of Sparse Positive Definite Matrices
Michał Kos, Krzysztof Podgórski, Hanqing Wu
Efficient inversion of large sparse positive definite matrices requires exploiting sparsity patterns beyond those captured by conventional bandwidth reduction. In this work, we rec…
math.ST2024
Unveiling low-dimensional patterns induced by convex non-differentiable regularizers
Ivan Hejný, Jonas Wallin, Małgorzata Bogdan +1
Popular regularizers with non-differentiable penalties, such as Lasso, Elastic Net, Generalized Lasso, or SLOPE, reduce the dimension of the parameter space by inducing sparsity or…
math.ST2019
On the asymptotic properties of SLOPE
Michał Kos, Małgorzata Bogdan
Sorted L-One Penalized Estimator (SLOPE) is a relatively new convex optimization procedure for selecting predictors in large data bases. Contrary to LASSO, SLOPE has been proved to…