A learning-enhanced projection method for solving convex feasibility\n problems

2019/05/01 by Janosch Rieger, Rieger, Janosch
Mathematics · Computer Science · Engineering · #Advanced Optimization Algorithms Research #Optimization and Variational Analysis #Sparse and Compressive Sensing Techniques

paper · pdf · doi:10.48550/arxiv.1905.00196

Abstract

We propose a generalization of the method of cyclic projections, which uses\nthe lengths of projection steps carried out in the past to learn about the\ngeometry of the problem and decides on this basis which projections to carry\nout in the future. We prove the convergence of this algorithm and illustrate\nits behavior in a first numerical study.\n

Related

• The method of cyclic intrepid projections: convergence analysis and numerical experiments
• Randomized projection methods for convex feasibility problems: conditioning and convergence rates
• Analysis of the convergence rate for the cyclic projection algorithm applied to basic semi-algebraic convex sets
• On the Effectiveness of Projection Methods for Convex Feasibility Problems with Linear Inequality Constraints
• Global convergence and acceleration of projection methods for feasibility problems involving union convex sets