Abstract
This paper modifies the Frank-Wolfe's algorithm. Under weaker conditions it proves that the modified algorithm is convergent, and specially under the assumption of convexity of the objective function that
without assuming {x k} is bounded.
Similar content being viewed by others
References
M. Frank and P. Wolfe. An Algorithm for Quadratic Programming.Naval Research Logistics Quarterly, 1959, 3: 95–110.
Wang Changyu. A Feasible Direction Method for Nonlinear Programming.Acta Mathematicae Sinica, 1982, 25(1): 15–19 (in Chinese).
Masso Fukushima. A Modified Frank-Wolfe Algorithm for Solving The Traffic Assignment Problem.Transpr. Res. B, 1984, 18B(2): 169–177.
M. Florian. An Improved Linear Approximation Algorithm for The Network Equilibrium Problem. Proc. of the IEEE Conf. on Decision & Control, 1977, 812–818.
C.C. Holloway. An Extension of Frank-Wolfe Method of Feasible Direction.Mathematical Programming, 1974, 6: 14–27.
M.J. Todd. On Convergence Properties of Algorithms for Unconstrained Minimization.IMA Journal of Numerical Analysis, 1989, 9(3): 435–441.
Author information
Authors and Affiliations
Additional information
This work is supported by Ecole Polytechnique of France and the National Natural Science Foundation of China.
Rights and permissions
About this article
Cite this article
Wu, F., Wu, S. A modified Frank-Wolfe algorithm and its convergence properties. Acta Mathematicae Applicatae Sinica 11, 285–291 (1995). https://doi.org/10.1007/BF02011194
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02011194