First-order methods of smooth convex optimization with inexact oracle
Tóm tắt
We introduce the notion of inexact first-order oracle and analyze the behavior of several first-order methods of smooth convex optimization used with such an oracle. This notion of inexact oracle naturally appears in the context of smoothing techniques, Moreau–Yosida regularization, Augmented Lagrangians and many other situations. We derive complexity estimates for primal, dual and fast gradient methods, and study in particular their dependence on the accuracy of the oracle and the desired accuracy of the objective function. We observe that the superiority of fast gradient methods over the classical ones is no longer absolute when an inexact oracle is used. We prove that, contrary to simple gradient schemes, fast gradient methods must necessarily suffer from error accumulation. Finally, we show that the notion of inexact oracle allows the application of first-order methods of smooth convex optimization to solve non-smooth or weakly smooth convex problems.
Tài liệu tham khảo
Baes, M.: Estimate Sequence Methods: Extensions and Approximations. IFOR Internal report, ETH Zurich, Switzerland (2009)
Correa, R., Lemarechal, C.: Convergence of some algorithms for convex minimization. Math. Program. Ser. A 62, 261–275 (1993)
D’Aspremont, A.: Smooth optimization with approximate gradient. SIAM J. Optim. 19, 1171–1183 (2008)
Devolder, O., Glineur, F., Nesterov, Y.: Double smoothing technique for large-scale linearly constrained convex optimization. SIAM J. Optim. 22(2), 702–727 (2012)
Hintermuller, M.: A proximal bundle method based on approximative subgradient. Comput. Optim. Appl. 20, 245–266 (2001)
Kiwiel, K.: A proximal bundel method with approximative subgradient linearization. SIAM J. Optim. 16, 1007–1023 (2006)
Khachiyan, L., Nemirovskii, A., Nesterov, Y.: Optimal methods of convex programming and polynomial methods of linear programming. In: Elster, H. (ed.) Modern Mathematical Methods of Optimization, pp. 75–115. Akademie Verlag, Berlin (1993)
Lan, G.: An optimal method for stochastic composite optimization. Math. Program. Ser. A 133(1–2), 365–397 (2012)
Nedic, A., Bertsekas, D.: The effect of deterministic noise in subgradient methods. Math. Program. Ser. A 125, 75–99 (2010)
Nemirovskii, A., Nesterov, Y.: Optimal methods for smooth convex minimization. Zh. Vichisl. Mat. Fiz. (In Russian) 25(3), 356–369 (1985)
Nemirovskii, A., Yudin, D.: Problem Complexity and Method Efficiency in Optimization. Wiley, New York (1983)
Nesterov, Y.: A method for unconstrained convex minimization with the rate of convergence of \({O}(\frac{1}{k^2})\). Doklady AN SSSR 269, 543–547 (1983)
Nesterov, Y.: On an approach to the construction of optimal methods of minimization of smooth convex function. Èkonom. i. Mat. Metody (In Russian) 24, 509–517 (1988)
Nesterov, Y.: Introductory Lectures on Convex Optimization: A Basic Course. Kluwer, Dordrecht (2004)
Nesterov, Y.: Smooth minimization of non-smooth functions. Math. Program. Ser. A 103, 127–152 (2005)
Nesterov, Y.: Excessive gap technique in nonsmooth convex minimization. SIAM J. Optim. 16, 235–249 (2005)
Nesterov, Y.: Smoothing technique and its applications in semidefinite optimization. Math. Program. A 110, 245–259 (2007)
Nesterov, Y.: Gradient methods for minimizing composite objective function. Math. Program. Ser. B. doi:10.1007/s10107-012-0629-5 (2012)
Polyak, B.T.: Introduction to Optimization. Optimization Software Inc, New York (1987)
Shor, N.Z.: Minimization Methods for Non-Differentiable Functions. Springer Series in Computational Mathematics. Springer, Berlin (1985)