Overview
pdNCG (primal-dual Newton Conjugate Gradients) is a MATLAB
implementation for the solution of unconstrained l1-regularized problems.
For example, Machine Learning problems, such as l1-regularized least-squares
and logistic regression, Compressed Sensing problems, such as l1-synthesis,
l1-analysis and isotropic total-variation. The solver is memoryless, it
requires only matrix-vector product operations, hence it is appropriate for
large-scale instances.
Download
- Source code:
Latest version (v3)
Previous version (v2)
Previous version (v1)
- Additional scripts: scripts that reproduce all experiments in the paper "A Preconditioner for a Primal-Dual Newton Conjugate Gradients Method for Compressed Sensing Problems" (Version 3 of pdNCG has been used for these experiments)
- Additional scripts: scripts that reproduce all experiments in the paper "A Second-Order Method for Compressed Sensing Problems with Coherent and Redundant Dictionaries" (Version 1 of pdNCG has been used for this experiment)
- Additional scripts: scripts that reproduce all experiments in the paper "A Second-Order Method for Strongly Convex L1-Regularization Problems" (version 2 of pdNCG has been used for these experiments)
License
pdNCG or primal-dual Newton Conjugate Gradients
Copyright (C) 2014, Ioannis Dassios, Kimon Fountoulakis, Robert Mansel Gower
and Jacek Gondzio
This program is free software: you can redistribute it and/or modify
it under the terms of the GNU General Public License as published by
the Free Software Foundation, either version 3 of the License, or
(at your option) any later version.
This program is distributed in the hope that it will be useful,
but WITHOUT ANY WARRANTY; without even the implied warranty of
MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the
GNU General Public License for more details.
References
-
K. Fountoulakis and J. Gondzio:
A Second-Order Method for Strongly-Convex L1-Regularization Problems,
Technical Report ERGO-14-005, School of Mathematics, 2014.
Published in Mathematical Programming.
-
I. Dassios, K. Fountoulakis and J. Gondzio:
A Second-Order Method for Compressed Sensing Problems with Coherent and Redundant Dictionaries,
Technical Report ERGO-14-007, School of Mathematics, 2014.
-
I. Dassios, K. Fountoulakis and J. Gondzio:
A Preconditioner for a Primal-Dual Newton Conjugate Gradients Method for Compressed Sensing Problems,
Technical Report ERGO-14-021, School of Mathematics, 2014.
Published in SIAM Journal on Scientific Computing.