A Data-Driven Parameter Prediction Method for HSS-Type Methods

Some matrix-splitting iterative methods for solving systems of linear equations contain parameters that need to be specified in advance, and the choice these directly affects efficiency corresponding methods. This paper uses a Bayesian inference-based Gaussian process regression (GPR) method predict relatively optimal some HSS-type iteration provide extensive numerical experiments compare prediction performance GPR with other existing Numerical results show using has advantage smaller computational effort, predicting more universality compared currently available finding