Abstract
We present a novel fully adaptive conditional gradient method with the step length regulation for solving pseudo-convex constrained optimization problems. We propose some deterministic rules of the step length regulation in a normalized direction. These rules guarantee to find the step length by utilizing the finite procedures and provide the strict relaxation of the objective function at each iteration. We prove that the sequence of the function values for the iterates generated by the algorithm converges globally to the objective function optimal value with sublinear rate.
Similar content being viewed by others
References
Dunn, J.C., Hafghbarger, S.: Conditional gradient algorithms with open loop step size rules. J. Math. Anal. Appl. 62, 432–444 (1978)
Lacoste-Julien, S., Jaggi, M., Schmidt, M., Pletscher, P.: Block-coordinate Frank–Wolfe optimization for structural SVMs. In: Proceedings of the 30th International Conference on Machine Learning, PMLR, vol. 28(1), pp. 53–61 (2013)
Braun, G., Pokutta, S., Zink, D.: Lazifying conditional gradient algorithms. arXiv preprint arXiv:1610.051220v4 (2018)
Lacoste-Julien, S.: Convergence rate of Frank–Wolfe for non-convex objectives. arXiv preprint arXiv:1607.00345 (2016)
Nesterov, Y.: Complexity bounds for primal-dual methods minimizing the model of objective function. Math. Program. 171(1–2), 311–330 (2018)
Bertsekas, D.P.: Nonlinear Programming, 2nd edn. Athena Scientific, Belmont (1999)
Yu, Y., Zhang, X., Schuurmans, D.: Generalized conditional gradient for sparse estimation. arXiv preprint arXiv:1410.4828v1 (2014)
Gabidullina, Z.R.: Relaxation methods with step regulation for solving constrained optimization problems. J. Math. Sci. 73(5), 538–543 (1995)
Mitchell, V.F., Dem’yanov, V.F., Malozemov, V.N.: Finding the point of polyhedron closest to origin. SIAM J. Control Optim. 12, 19–26 (1974)
Gabidullina, Z.R.: The problem of projecting the origin of euclidean space onto the convex polyhedron. Lobachevskii J. Math. 39(1), 35–45 (2018)
Gabidullina, Z.R.: Solving of a projection problem for convex polyhedra given by a system of linear constraints. In: Constructive Nonsmooth Analysis and Related Topics (Dedicated to the Memory of V.F. Demyanov), CNSA Proceedings—IEEE, Art. 7973958 (2017). http://ieeexplore.ieee.org/abstract/document/7973958/
Gabidullina, Z.R.: The Minkowski difference for convex polyhedra and some its applications. arXiv preprint arXiv:1903.03590 (2019)
Mangasarian, O.L.: Pseudo-convex functions. J. Soc. Ind. Appl. Math. Ser. A Control 3, 281–290 (1965)
Gabidullina, Z.R.: Convergence of the constrained gradient method for a class of nonconvex functions. J. Sov. Math. 50(5), 1803–1809 (1990)
Gabidullina, Z.R.: Adaptive methods with step length regulation for solving pseudo-convex programming problems. Dissertation for the Degree of Candidate of Science in Physics and Mathematics, Kazan (1994)
Vasil’ev, F.P.: Numerical Methods for Solving Extremum Problems. Nauka, Moscow (1980)
Reklaitis, G.V., Ravindran, A., Ragsdell, K.M.: Engineering Optimization: Methods and Applications, 2nd edn. Wiley, Hoboken (2006)
Beltukov, I.B., Shurygina, M.N.: A study of one adaptive method for mathematical programming. In: Optimization Methods and Applications, pp. 5–13. Irkutsk (1988) (in Russian)
Acknowledgements
The author thanks the anonymous referees and the editor for their helpful comments and remarks on a previous version of the paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Alexanre Cabot.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Gabidullina, Z.R. Adaptive Conditional Gradient Method. J Optim Theory Appl 183, 1077–1098 (2019). https://doi.org/10.1007/s10957-019-01585-w
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10957-019-01585-w
Keywords
- Optimization problems
- Pseudo-convex function
- Adaptation
- Descent direction
- Normalization
- Step length
- Regulation
- Rate of convergence