A finitely convergent ``row-action'' method for the convex feasibility problem (Q1109529)

The authors present a modification of the cyclic subgradient projection (MCSP) method by \textit{Y. Censor} and \textit{A. Lenz} [Math. Programming 24, 233-235 (1982; Zbl 0491.90077)] which solves the convex feasibility problem in a finite number of steps when a Slater type condition holds. The MCSP method is a row-action method useful for handling huge and sparse systems of convex inequalities. A linear rate of convergence for the cyclic subgradient projection method is proved assuming the presence of the Slater condition, too.