Parallel iterative regularization methods for solving systems of ill-posed equations
Introduction
Various problems of science and engineering can be reduced to finding a solution of a given simultaneous system of operator equationswhere are given possibly nonlinear operators in a real Hilbert space H, and is an unknown element.
The well-known convex feasibility problem, appearing in many areas, such as optimization theory, image processing, and radiation therapy treatment planning, consists of computing a common solution of equationswhere are projection operators onto given closed convex sets , . Problem (1.2) is usually referred to as a common fixed point problem.
A lot of sequential and parallel algorithms for the convex feasibility and common fixed point problems have been proposed. The cyclic projection algorithm, the block-iterative projection algorithm and the explicit iteration algorithm, to name only a few (see, e.g., [4], [5], [7], [8]).
In what follows, we are interested in parallel methods for solving (1.1), where each equation is ill-posed. We refer to [1], [2], [3], [7], [9], [11], [13] for the general theory of ill-posed problems, and particularly, to [1], [2], [3], [7] for iterative regularization methods. Unfortunately, these methods, except that one in [7], cannot be used directly for finding a common solution of ill-posed equations. Besides, no numerical examples are available in [7].
The present work is motivated by an interesting idea on regularization for systems of nonlinear equations involving monotone operators [7], and the parallel splitting-up technique for solving nonlinear elliptic problems [10]. We propose a parallel implicit iterative regularization method (PIIRM) and a parallel explicit iterative regularization method (PEIRM) for solving system (1.1). Moreover, the explicit regularization method suggested in [7] is a particular case of our PEIRM.
An outline of the remainder of the paper is as follows: In Section 1 we recall some definitions and results that will be used later on in the proof of our main theorems. Section 2 presents our main convergence results, and finally, in Section 3 some applications of the proposed parallel iterative regularization methods are discussed. Definition 1.1 An operator is said to be demiclosed at u if whenever and , then .
Hereafter, the symbols ⇀ and → denote the weak and the convergence in norm, respectively. It is well known (see [6]), that if is a nonexpansive mapping in a uniformly convex Banach space E, and Ω is a nonempty convex subset containing zero of E, then is demiclosed at zero. Definition 1.2 An operator is called – inverse-strongly monotone, ifwhere c is some positive constant.
It is proved in [9], that every linear self-adjoint, nonnegative and compact operator in a Hilbert space is inverse-strongly monotone, and the difference between the identity operator and a nonexpansive mapping is an inverse-strongly monotone operator.
Obviously, every inverse-strongly monotone operator is demiclosed at any point. Indeed, let and . From the inverse-strong monotonicity of A, it followsThe last sum tends to zero because in the first term, , while is bounded and in the second one, . Thus, , hence A is demiclosed at x. We also note that every inverse-strongly monotone operator is monotone and not necessarily strongly monotone.
For further consideration, we need the following lemma (see [12]). Lemma 1.1 Let and be sequences of nonnegative numbers, be a sequence of positive numbers, satisfying the inequalitieswhere and . Then . Theorem 1.2 Let be - inverse-strongly monotone operators, defined on the whole space H. Let be a converging to zero sequence of positive numbers. Consider a regularized equation for system (1.1)Then the following statements hold: For each , the regularized equation (1.3) has a unique solution . The regularized solution converges to the minimal-norm solution of system (1.1). Moreover, there hold estimates:and
Proof
See [7, Theorem 2.1]. □
We note that the demiclosedness of operators has been used in the proof of Theorem 1.2.
Section snippets
Convergence results
In what follows, we assume that system (1.1) always possesses a solution. Our idea is to solve the regularized equation (1.3) by a modified parallel splitting-up algorithm [10]. The modification is that for each , only one iteration is performed in solving (1.3). Theorem 2.1 Let and be two sequences of positive numbers, such that , , as and . Suppose the nth-approximation is found ( is given). Then the following parallel iterative regularization
Applications
We begin with the so-called image reconstruction problem, consisting of finding the weights x knowing the views and the projections , such thatProblem (3.1) can be written in the form (1.1) with , where is the orthogonal projection of x onto the hyperplane . It is well known (see [4], [5], [7]), that and are firmly nonexpansive, or 1-inverse-strongly monotone. A simple calculation shows that Eq. (2.1) in
Conclusion
Based on the regularization idea for a system of operator equations and the parallel splitting-up technique, we have proposed two parallel iterative regularization methods for solving systems of operator equations. Their better performance towards the explicit iterative algorithm suggested by Buong and Son [7] have been examined on some examples. The convergence of the proposed methods in the inexact data case as well as their finite dimensional approximations may be analyzed similarly as that
Acknowledgements
The authors would like to express their special thanks to the referees, whose careful reading and many constructive comments led to a considerable improvement of the paper. The work is partially supported by the Vietnam National Foundation for Science and Technology Development.
References (13)
Viscosity approximation methods for nonexpansive mappings
Journal of Mathematical Analysis and Applications
(2004)- et al.
Ill-Posed Problems: Theory and Applications
(1994) - et al.
Iterative Methods for Solving Ill-Posed Problems
(1989) - et al.
Iterative Regularization Methods for Nonlinear Ill-Posed Problems
(2008) - et al.
On projection algorithms for solving convex feasibility problems
SIAM Review
(1996) - et al.
The method of cyclic projections for closed convex sets in Hilbert space
Contemporary Mathematics
(1997)
Cited by (10)
Parallel iterative regularization algorithms for large overdetermined linear systems
2010, International Journal of Computational MethodsApproximate solution for integral equations involving linear Toeplitz plus Hankel parts
2021, Computational and Applied MathematicsExplicit Extragradient-Like Method with Regularization for Variational Inequalities
2019, Results in MathematicsOn the numerical solution of system of linear algebraic equations with positive definite symmetric ill-posed matrices
2018, WSEAS Transactions on MathematicsRegularization for the problem of finding a solution of a system of nonlinear monotone ill-posed equations in banach spaces
2018, Journal of the Korean Mathematical SocietyParallel methods for regularizing systems of equations involving accretive operators
2014, Applicable Analysis