Parallel hybrid methods for generalized equilibrium problems and asymptotically strictly pseudocontractive mappings (Q513526): Difference between revisions

The article deals with the generalized equilibrium problem in a Hilbert space \(H\). More precisely, let \(C\) be a nonempty closed subset of a Hilbert space \(H\) and \(\{f_i\}_{i=1}^N\) is a finite family of bifunction with the following properties: (A1) \(f(x,x) = 0\); (A2) \(f(x,y) + f(y,x) \leq 0\); (A3) \(\limsup\limits_{t \to 0}\, f(tz + (1 - t)x,y) \leq f(x,y)\); (A4) \(f(x,\cdot)\) (\(x \in C\)) are convex and lower semicontinuous; \(\{A_i\}_{i=1}^N\) a finite family of \(\alpha\)-inverse strongly monotone mappings; and \(\{S_j\}_{j=1}^M\) a finite family of asymptotically \(\kappa\)-strictly pseudocontractive mappings with the characteristic sequence \(\{k_n\} \subset [1;+\infty)\), \(k_n \to 1\). It is assumed that the set \[ F = \bigg(\bigcap_{i=1}^N \text{GEP}\, (f_i,A_i)\bigg) \cap \bigg(\bigcap_{j=1}^M \text{Fix}\, (S_j)\bigg) \] is nonempty and bounded (here \(\text{GEP}\, (f,A)\) is a problem of finding of \(x\) such that \[ x \in C:\;f(x,y) + \langle Ax,y - x \rangle \geq 0 \quad (y \in C)). \] The problem under consideration is the approximate construction of the operator \[ P_Fx = \text{arg \;min}\, \{\|y - x\|:\;y \in F\}. \] The author offers two parallel hybrid algorithms for the calculation of sequences \(\{x_n\}\), \(\{u_n^i\}\), and \(\{z_n^j\}\) that, under some standard assumptions, converge strongly to \(P_Fx_0\). Some numerical examples illustrate the efficiency of the proposed iterative methods.