Note on the modified relaxation CQ algorithm for the split feasibility problem (Q1653279): Difference between revisions

Let two Hilbert spaces $H_{1}$ and $H_{2}$, a bounded linear operator $A:H_{1}\rightarrow H_{2}$ and two nonempty, closed and convex sets $C\subseteq H_{1}$ and $Q\subseteq H_{2}$ be given. The split feasibility problem (SFP) is to find $x^{\ast}\in C$ such that $Ax^{\ast}\in Q$. The paper under review studies a modified relaxation CQ algorithm with Armijo-line search. It is shown that the sequence generated by the algorithm weakly converges to a solution of the SFP, under the following assumptions: (a) the set of solutions is not empty, (b) the sets $C$ and $Q$ are defined as $C=\left \{ x\in H_{1}|c(x)\leq0\right \}$, $Q=\left \{ x\in H_{2}|q(x)\leq0\right \} $ where the functions $c:H_{1}\rightarrow \mathbb{R}$ and $q:H_{2} \rightarrow \mathbb{R}$ are convex, and (c) at each $x\in H_{1}$, $y\in H_{2}$, at least one subgradient $\xi \in \partial c(x)$ and $\eta \in \partial q(x)$ can be calculated. \par The result extends an analogous result by \textit{B. Qu} and \textit{N. Xiu} [Inverse Probl. 21, No. 5, 1655--1665 (2005; Zbl 1080.65033)] to infinite-dimensional Hilbert spaces. \par In a numerical example, the algorithm above is compared to the relaxed CQ algorithm, proposed by Q. Yang.