Abstract
We provide Frank–Wolfe (\(\equiv \) Conditional Gradients) method with a convergence analysis allowing to approach a primal-dual solution of convex optimization problem with composite objective function. Additional properties of complementary part of the objective (strong convexity) significantly accelerate the scheme. We also justify a new variant of this method, which can be seen as a trust-region scheme applying to the linear model of objective function. For this variant, we prove also the rate of convergence for the total variation of linear model of composite objective over the feasible set. Our analysis works also for quadratic model, allowing to justify the global rate of convergence for a new second-order method as applied to a composite objective function. To the best of our knowledge, this is the first trust-region scheme supported by the worst-case complexity analysis both for the functional gap and for the variation of local quadratic model over the feasible set.
Similar content being viewed by others
References
Conn, A.B., Gould, N.I.M.: Toint, PhL: Trust Region Methods. SIAM, Philadelphia (2000)
Dunn, J.: Convergence rates for conditional gradient sequences generated by implicit step length rules. SIAM J. Control Optim. 18(5), 473–487 (1980)
Frank, M., Wolfe, P.: An algorithm for quadratic programming. Nav. Res. Logist. Q. 3, 149–154 (1956)
Freund, R.M., Grigas, P.: New analysis and results for the Frank-Wolfe method. Math. Program. (2014). doi:10.1007/s10107-014-0841-6
Garber, D., Hazan, E.: A linearly convergent conditional gradient algorithm with application to online and stochastic optimization (2013). arXiv:1301.4666v5
Garber, D., Hazan, E.: Faster rates for the Frank–Wolfe method over strongly-convex sets (2015). arXiv:1406.1305v2
Harchaoui, Z., Juditsky, A., Nemirovski, A.: Conditional gradient algorithms for norm-regularized smooth convex optimization. Math. Program. (2014). doi:10.1007/s10107-014-0778-9
Jaggi, M.: Revisiting Frank–Wolfe: projection-free sparse convex optimization. In: Proceedings of the 30th International Conference on Machine Learning, Atlanta, Georgia (2013)
Juditsky, A., Nemirovski, A.: Solving variational inequalities with monotone operators on domains given by linear minimization oracles. Math. Program. Ser. A 156, 221–256 (2015). doi:10.1007/s10107-015-0876-3
Lacoste-Julien, S., Jaggi, M., Schmidt, M., Pletscher, P.: Block-coordinate Frank–Wolfe optimization of structural svms. In: Proceedings of the 30th International Conference on Machine Learning, Atlanta, Georgia (2013)
Lan, G.: The complexity of large-scale convex programming under a linear optimization oracle (2014). arXiv:1309.5550v2
Migdalas, A.: A regularization of the Frank-Wolfe method and unification of certain nonlinear programming methods. Math. Program. 63, 331–345 (1994)
Nesterov, Y.: Introductory Lectures on Convex Optimization. Kluwer, Boston (2004)
Nesterov, Y.: Primal-dual subgradient methods for convex problems. Math. Program. 120(1), 221–259 (2009)
Nesterov, Y.: Gradient methods for minimizing composite functions. Math. Program. 140(1), 125–161 (2013)
Nesterov, Y.: Universal gradient methods for convex optimization problems. Math. Program. (2014). doi:10.1007/s10107-014-0790-0
Nesterov, Y., Nemirovskii, A.: Interior-Point Polynomial Algorithms in Convex Programming. SIAM, Philadelphia (1994)
Nesterov, Y., Polyak, B.: Cubic regularization of Newton’s method and its global performance. Math. Program. 108(1), 177–205 (2006)
Nesterov, Y., Shikhman, V.: Quasi-monotone subgradient methods for nonsmooth convex minimization. JOTA (2014). doi:10.1007/s10957-014-0677-5
Yu, Y., Zhang, X., Schuurmans, D.: Generalized conditional gradient for sparse estimation (2014). arXiv:1410.4828 (Under review in Journal of Machine Learning Research)
Acknowledgements
The author is very thankful to both anonymous reviewers for their valuable comments.
Author information
Authors and Affiliations
Corresponding author
Additional information
Scientific results of this paper were obtained with support of RSF Grant 17-11-01927.
Rights and permissions
About this article
Cite this article
Nesterov, Y. Complexity bounds for primal-dual methods minimizing the model of objective function. Math. Program. 171, 311–330 (2018). https://doi.org/10.1007/s10107-017-1188-6
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10107-017-1188-6
Keywords
- Convex optimization
- Complexity bounds
- Linear optimization oracle
- Conditional gradient method
- Trust-region method