RCGLS replaces the gradient in CGLS with a randomized coordinate version via a constraint correction view, proving linear convergence in expectation better than randomized coordinate descent, plus sparse implementation and ridge regression extension.
Connecting rand omized iterative methods with Krylov sub- spaces
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
fields
math.NA 3years
2026 3verdicts
UNVERDICTED 3representative citing papers
The Gearhart-Koshy acceleration yields linear convergence to the least-norm solution for tensor linear systems with improved rates over plain Kaczmarz across incremental, shuffle-once, and random-reshuffling schemes.
Refines subspace preconditioning for randomized linear solvers via QR-like factorization, enabling implicit use and proving expected linear convergence while reducing to a smaller system with good singular values.
citing papers explorer
-
Randomized conjugate gradient least squares
RCGLS replaces the gradient in CGLS with a randomized coordinate version via a constraint correction view, proving linear convergence in expectation better than randomized coordinate descent, plus sparse implementation and ridge regression extension.
-
Linear convergence of Gearhart-Koshy accelerated Kaczmarz methods for tensor linear systems
The Gearhart-Koshy acceleration yields linear convergence to the least-norm solution for tensor linear systems with improved rates over plain Kaczmarz across incremental, shuffle-once, and random-reshuffling schemes.
-
On subspace-constrained preconditioning for randomized iterative methods
Refines subspace preconditioning for randomized linear solvers via QR-like factorization, enabling implicit use and proving expected linear convergence while reducing to a smaller system with good singular values.