We study a class of regularization methods for solving least-squares ill-posed problem with a convex constraint. Convergence and convergence rate results are proven for the problems which satisfy so called power source condition. All the results are obtained under the assumptions that, instead of exact initial data, only their approximations are known.
1 April 2024 in Computational Mathematics and Mathematical Physics
The objective of this manuscript is to study the convergence of three-step approximation methods for quasi-variational inequalities in the general case. First, we propose the three-step dynamical system and carry out an asymptotic analysis for the generated trajectories. The explicit time discretization of this system results into a three-step iterative method, which we prove to converge also when it is applied to strongly-monotone quasi-variational inequalities. In addition, we show that linear convergence is guaranteed under strong-monotonicity.