On the Rate of Convergence of a Partially Asynchronous Gradient Projection Algorithm

Recently, Bertsekas and Tsitsiklis proposed a partially asynchronous implementation of the gradient projection algorithm of Goldstein and Levitin and Polyak for the problem of minimizing a differentiable function over a closed convex set. In this paper, the rate of convergence of this algorithm is analyzed. It is shown that if the standard assumptions hold (that is, the solution set is nonempty and the gradient of the function is Lipschitz continuous) and (i) the isocost surfaces of the objective function, restricted to the solution set, are properly separated and (ii) a certain multifunction associated with the problem is locally upper Lipschitzian, then this algorithm attains a linear rate of convergence.

[1]  John N. Tsitsiklis,et al.  Distributed Asynchronous Deterministic and Stochastic Gradient Optimization Algorithms , 1984, 1984 American Control Conference.

[2]  Dimitri P. Bertsekas,et al.  Constrained Optimization and Lagrange Multiplier Methods , 1982 .

[3]  J. J. Moré.  Gradient projection techniques for large-scale optimization problems , 1989, Proceedings of the 28th IEEE Conference on Decision and Control,.

[4]  Boris Polyak,et al.  Constrained minimization methods , 1966 .

[5]  E. A. Nurminskii.  Convergence of the gradient projection method , 1973 .

[6]  J. Dunn.  On the convergence of projected gradient processes to singular critical points , 1987 .

[7]  Paul H. Calamai,et al.  Projected gradient methods for linearly constrained problems , 1987, Math. Program..

[8]  D. Bertsekas.  Projected Newton methods for optimization problems with simple constraints , 1981, 1981 20th IEEE Conference on Decision and Control including the Symposium on Adaptive Processes.

[9]  Y. C. Cheng.  On the gradient-projection method for solving the nonsymmetric linear complementarity problem , 1984 .

[10]  P. Tseng,et al.  On the convergence of the coordinate descent method for convex differentiable minimization , 1992 .

[11]  S. M. Robinson.  Generalized equations and their solutions, part II: Applications to nonlinear programming , 1982 .

[12]  Jong-Shi Pang,et al.  A Posteriori Error Bounds for the Linearly-Constrained Variational Inequality Problem , 1987, Math. Oper. Res..

[13]  A. Goldstein.  Convex programming in Hilbert space , 1964 .

[14]  D. Bertsekas,et al.  TWO-METRIC PROJECTION METHODS FOR CONSTRAINED OPTIMIZATION* , 1984 .

[15]  S. M. Robinson.  Some continuity properties of polyhedral multifunctions , 1981 .

[16]  John N. Tsitsiklis,et al.  Parallel and distributed computation , 1989 .

[17]  John N. Tsitsiklis,et al.  Problems in decentralized decision making and computation , 1984 .

[18]  Paul Tseng,et al.  Error Bound and Convergence Analysis of Matrix Splitting Algorithms for the Affine Variational Inequality Problem , 1992, SIAM J. Optim..

[19]  P. Tseng,et al.  On the linear convergence of descent methods for convex essentially smooth minimization , 1992 .

[20]  D. Bertsekas.  On the Goldstein-Levitin-Polyak gradient projection method , 1974, CDC 1974.

[21]  D. Bertsekas,et al.  Distributed asynchronous optimal routing in data networks , 1984, The 23rd IEEE Conference on Decision and Control.

[22]  J. Dunn.  Global and Asymptotic Convergence Rate Estimates for a Class of Projected Gradient Processes , 1981 .