Solving A System Of Linear Equations By Randomized Orthogonal Projections
share
Randomization has shown catalyzing effects in linear algebra with promising perspectives for tackling computational challenges in large-scale problems. For solving a system of linear equations, we demonstrate the convergence of a broad class of algorithms that at each step pick a subset of n equations at random and update the iterate with the orthogonal projection to the subspace those equations represent. We identify, in this context, a specific degree-n polynomial that non-linearly transforms the singular values of the system towards equalization. This transformation to singular values and the corresponding condition number then characterizes the expected convergence rate of iterations. As a consequence, our results specify the convergence rate of the stochastic gradient descent algorithm, in terms of the mini-batch size n, when used for solving systems of linear equations.
READ FULL TEXT