Decentralized ADMM for factorization-based Low-rank matrix estimation
Zihao Song1, Weihua Zhao1, Rui Li2
1School of Mathematics and Statistics, Nantong University, Nantong, China.
None:
We consider the problem of recovering a low-rank matrix in a distributed setting, based on a convex loss function and non-convex matrix factorization. We use a linearized and decentralized alternating direction method of multipliers (ADMM) algorithm to compute the consensus solution. We establish local linear convergence (up to the approximation error when the unconstrained solution is not exactly low-rank) of the method despite the optimization problem being non-convex due to the factorization. Numerical examples are presented to illustrate the performance.
Related Concept Videos
Gaussian Elimination: Problem Solving
Partial Fractions
Compacting Factor test
The procedure begins by placing concrete into the upper hopper without any compaction. Once filled, the bottom door of this hopper is opened,...
One-Compartment Open Model: Wagner-Nelson and Loo Riegelman Method for ka Estimation
On...
Mechanistic Models: Compartment Models in Algorithms for Numerical Problem Solving
In individual population analyses, different algorithms are employed, such as Cauchy's method, which uses a...
Extraction: Partition and Distribution Coefficients
For extracting a solute from an aqueous phase into an organic...
