A fast algorithm for matrix balancing
The Sinkhorn-Knopp algorithm [\textit{R. Sinkhorn} and \textit{P. Knopp}, Pac. J. Math. 21, 343--348 (1967; Zbl 0152.01403)] offers an approach to balancing matrices but it does not work well for sparse matrices. In this paper, the authors derive some new algorithms that are fast for large matrix-balancing problems. They show that an inexact Newton method offers a simple way of overcoming the limitation of the Sinkhorn-Knopp algorithm. They make sure that if they remain in the positive cone then they show improvements for both symmetric and nonsymmetric matrices. They point out that if the diagonal of the matrix contain widely varying elements then the proposed algorithm converges slowly.
- A Comparative Study of Algorithms for Matrix Balancing
- On the Complexity of Matrix Balancing
- Vector and parallel computing for matrix balancing
- Balancing an arbitrary matrix using computations with Stieltjes matrices
- Matrix Balancing on a Massively Parallel Connection Machine
- Interval-constrained matrix balancing
- scientific article; zbMATH DE number 7376020
- Balancing sparse matrices for computing eigenvalues
- scientific article; zbMATH DE number 2096649
- Near-linear convergence of the random Osborne algorithm for matrix balancing
- Lagrangian numerical methods for ocean biogeochemical simulations
- Scaling matrices and counting the perfect matchings in graphs
- A note on overrelaxation in the Sinkhorn algorithm
- Accelerating the Sinkhorn-Knopp iteration by Arnoldi-type methods
- An empirical Bayes approach for the identification of long-range chromosomal interaction from Hi-C data
- Network modelling of topological domains using Hi-C data
- Stochastic matrix-free equilibration
- Balancing an arbitrary matrix using computations with Stieltjes matrices
- Analysis of a classical matrix preconditioning algorithm
- Boosting of image denoising algorithms
- Arbitrarily regularizable graphs
- A Comparative Study of Algorithms for Matrix Balancing
- The Sinkhorn–Knopp Algorithm: Convergence and Applications
- Matrix Balancing on a Massively Parallel Connection Machine
- On the Complexity of Matrix Balancing
- Semidual regularized optimal transport
- Matrix balancing in \(L_p\) norms: bounding the convergence rate of Osborne's iteration
- scientific article; zbMATH DE number 7376020 (Why is no real title available?)
- Constructive quantum scaling of unitary matrices
- Preconditioning techniques based on the Birkhoff-von Neumann decomposition
- Better and simpler error analysis of the Sinkhorn-Knopp algorithm for matrix scaling
- MATRIX BALANCING PROBLEM AND BINARY AHP(<Special Issue>the 50th Anniversary of the Operations Research Society of Japan)
- scientific article; zbMATH DE number 7651215 (Why is no real title available?)
- An entropy optimizing RAS-equivalent algorithm for iterative matrix balancing
- Translocation Detection from Hi-C Data via Scan Statistics
- Introducing the Class of SemiDoubly Stochastic Matrices: A Novel Scaling Approach for Rectangular Matrices
- Matrix Balancing Based Interior Point Methods for Point Set Matching Problems
- Numerical simulation of the two-dimensional fractional Schrödinger equation for describing the quantum dynamics on a comb with the absorbing boundary conditions
- Statistical curve models for inferring 3D chromatin architecture
- The mechanism analysis for the fractional quantum dynamics on a comb structure with the absorbing boundary conditions
- On Sinkhorn's DAD theorem and the self-consistency equation in COSMO-based activity coefficient models
- Scalable approximate optimal diagonal preconditioning
- Algorithms for symmetric Birkhoff-von Neumann decomposition of symmetric doubly stochastic matrices
- Model-based spatial reconstruction of large-scale biomolecules via Bayesian inference of a hierarchical spatial model
- Interval-constrained matrix balancing
- Numerical algorithm for finding balanced metrics
This page was built for publication: A fast algorithm for matrix balancing
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2841064)
