Abstract:The Kaczmarz algorithm is a simple iterative scheme for solving consistent linear systems. At each step, the method projects the current iterate onto the solution space of a single constraint. Hence, it requires low cost per iteration and storage, and it has a linear rate of convergence. Distributed implementations of Kaczmarz have recently become the de facto architectural choice for large-scale linear systems. Therefore, in this paper we develop a family of randomized block Kaczmarz algorithms that uses at each step a subset of the constraints and extrapolated stepsizes, and can be deployed on distributed computing units. Our approach is based on several new ideas and tools, including stochastic selection rules for the blocks of rows, stochastic conditioning of linear systems, and novel strategies for designing extrapolated stepsizes. We prove that randomized block Kaczmarz algorithms converge linearly in expectation, with a rate depending on the geometric properties of the matrix and its submatrices and on the size of the blocks. Our convergence analysis reveals that the algorithm is most effective when it is given a good sampling of the rows into well-conditioned blocks. Besides providing a general framework for the design and analysis of randomized block Kaczmarz methods, our results resolve an open problem in the literature related to the theoretical understanding of observed practical efficiency of extrapolated block Kaczmarz methods. We also propose an accelerated block Kaczmarz scheme, that is, acceleration in the sense of Chebyshev semi-iterative methods, where the stepsize is chosen based on the roots of Chebyshev polynomials, and we derive convergence rates depending on the square root of the geometric properties of the matrix. Finally, numerical examples illustrate the benefits of the new algorithms.

On the Error Estimate of the Randomized Double Block Kaczmarz Method

Error estimates of Kaczmarz and randomized Kaczmarz methods

An Accelerated Block Randomized Kaczmarz Method

On the alternating randomized block Kaczmarz method

On the convergence analysis of the greedy randomized Kaczmarz method

Refined upper bounds for the convergence of the randomized extended Kaczmarz and Gauss-Seidel algorithms

A quantile-based block Kaczmarz algorithm for solving large consistent linear systems

On global randomized block Kaczmarz algorithm for solving large-scale matrix equations

Randomized Kaczmarz iteration methods: Algorithmic extensions and convergence theory

On Multi-Step Randomized Extended Kaczmarz Method for Solving Large Sparse Inconsistent Linear Systems

On the Randomized Kaczmarz Algorithm

Improved Analyses of the Randomized Power Method and Block Lanczos Method

Preasymptotic convergence of randomized Kaczmarz method

Faster Randomized Block Kaczmarz Algorithms

Randomized Kaczmarz method with adaptive stepsizes for inconsistent linear systems

Randomized block Kaczmarz method with projection for solving least squares

Efficient randomized block Kaczmarz method for linear feasibility

On Convergence Rates of Kaczmarz-Type Methods with Different Selection Rules of Working Rows

Randomized Block Kaczmarz Methods for Inner Inverses of a Matrix

Randomized Kaczmarz algorithm with averaging and block projection