Abstract
In this paper we propose a modification of the mirror descent method for non-smooth stochastic convex optimization problems on the unit simplex. The optimization problems considered differ from the classical ones by availability of function values realizations. Our purpose is to derive the convergence rate of the method proposed and to determine the level of noise that does not significantly affect the convergence rate.
Similar content being viewed by others
References
Nesterov, Yu.E., Algorithmic Models of Human Behavior. A Presentation at the Mathematical Seminar, Moscow: MFTI&MTSNMO, September 14, 2012 (http://www.mathnet.ru/php/seminars. phtml?option lang=rus&presentid=6990).
Duchi, J.C., Jordan, M.I., Wainwright, M.J., and Wibisono, A., Optimal Rates for Zero-order Convex Optimization: The Power of Two Function Evaluations, IEEE Transact. Inform., 2015, vol. 61, no. 5, pp. 2788–2806 (http://www.eecs.berkeley.edu/~wainwrig/Papers/DucZero15.pdf).
Agarwal, A., Bartlett, P.L., Ravikumar, P., and Wainwright, M.J., Information-theoretic Lower Bounds on the Oracle Complexity of Stochastic Convex Optimization, IEEE Transact. Inform., 2012, vol. 58, no. 5, pp. 3235–3249, arXiv:1009.0571.
Kiefer, J. and Wolfowitz, J., Statistical Estimation on the Maximum of a Regression Function, Ann. Math. Statist., 1952, vol. 23, pp. 462–466.
Polyak, B.T., Vvedenie v optimizatsiyu, Moscow: Nauka, 1983. Translated into English under the title Introduction to Optimization, New York: Optimization Software, 1987.
Granichin, O.N. and Polyak, B.T., Randomizirovannye algoritmy otsenivaniya i optimizatsii pri pochti proizvol’nykh pomekhakh (Randomized Algorithms of Estimation and Optimization under Almost Arbitrary Noise) Moscow: Nauka, 2003.
Nemirovskii, A.S. and Yudin, D.B., Slozhnost’ zadach i effektivnost’ metodov optimizatsii (Complexity of Problems and Efficiency of the Optimization Methods), Moscow: Nauka, 1979.
Konecný, J. and Richárik, P., Simple Complexity Analysis of Simplified Direct Search, in e-print, 2014, arXiv:1410.0390.
Shapiro, A., Dentcheva, D., and Ruszczynski, A., Lecture on Stochastic Programming. Modeling and Theory, in MPS-SIAM Series Optim., 2014.
Nesterov, Y., Primal-Dual Subgradient Methods For Convex Problems, Math. Program., Ser. B, 2009, vol. 120(1), pp. 261–283.
Yuditskii, A.B., Nazin, A.V., Tsybakov, A.B., and Vayatis, N., Recurrent Aggregation of Estimates by the Method of Mirror Descent with Averaging, Probl. Peredachi Inf., 2005, vol. 41, no. 4, pp. 78–96.
Nemirovski, A., Juditsky, A., Lan, G., and Shapiro, A., Stochastic Approximation Approach to Stochastic Programming, SIAM J. Optim., 2009, vol. 19, no. 4, pp. 1574–1609.
Gasnikov, A.V., Nesterov, Yu.E., and Spokoinyi, V.G., On Efficiency of One Method for Randomization of the Mirror Descent in the Problems of Online Optimization, Zh. Vychisl. Mat. Mat. Fiz., 2015, vol. 55, no. 4, pp. 55–71.
Gasnikov, A.V., Dvurechenskii, P.E., and Nesterov, Yu.E., Stochastic Gradient Methods with Inaccurate Oracle, Tr. MFTI, 2016, vol. 8, no, 1. pp. 41–91, arxiv:1411.4218.
Bogolubsky, L., Dvurechensky, P., Gasnikov, A., et al., Learning Supervised PageRank with Gradientbased and Gradient-free Optimization Methods, in e-print, 2016, arXiv:1603.00717.
Gasnikov, A.V., Dvurechenskii, P.E., and Kamzolov, D.I., Gradient and Direct Methods with Inaccurate Oracle for Problems of Stochastic Optimization, in Dinamika sistem and protsessy upravleniya (System Dynamics and Control Processes), Proc. Int. Conf. in honour of 90th Anniversary of Academician N.N. Krasovskii, Yekaterinburg, September 15–20, 2014, Yekaterinburg: Krasovskii Inst. Math. Mech. Uro RAN, 2015, pp. 111–117, arXiv:1502.06259.
Belloni, A., Liang, T., Narayanan, H., and Rakhlin, A., Escaping the Local Minima via Simulated Annealing: Optimization of Approximately Convex Functions, JMLR. Workshop Conf. Proc., 2015, vol. 40, pp. 1–26, arXiv:1501.07242.
Nesterov, Yu., Random Gradient-free Minimization of Convex Functions, in CORE Discussion Paper 2011/1, 2011.
Spall, J.C., Introduction to Stochastic Search and Optimization: Estimation, Simulation and Control, New York: Wiley, 2003.
Bubeck, S. and Cesa-Bianchi, N., Regret Analysis of Stochastic and Nonstochastic Multi-armed Bandit Problems, Foundat. Trends Machine Learning, 2012, vol. 5, no. 1, pp. 1–122, arXiv:1204.5721.
Nemirovski, A., Lectures on Modern Convex Optimization Analysis, Algorithms, and Engineering Applications, Philadelphia: SIAM, 2013 (http://www2.isye.gatech.edu/~nemirovs/Lect ModConvOpt.pdf).
Leadbetter, M., Lindgren, G., and Rootzen, K., Extremes of Related Properties of Random Sequences and Processes, New York: Springer-Verlag, 1986. Translated under the title Ekstremumy sluchainykh posledovatel’nostei i protsessov, Moscow: Mir, 1989.
Ledoux, M., Concentration of Measure Phenomenon, Providence: Am. Math. Soc. (Math. Surveys Monogr., vol. 89), 2001.
Boucheron, S., Lugoshi, G., and Massart, P., Concentration Inequalities: A Nonasymptotic Theory of Independence, Oxford: Oxford Univ. Press, 2013.
Gasnikov, A.V., Dvurechenskii, P.E., Dorn, Yu.V., and Maksimov, Yu.V., Numerical Methods for Seeking Equilibrium Flow Distribution in the Backman Model and Stable Dynammice Model, Mat. Modelirovanie, 2016, vol. 28 (in press), arXiv:1506.00293.
Ibragimov, I.A. and Khas’minskii, R.Z., Asimptoticheskaya teoriya otsenivaniya (Asymptotic Estimation Theory), Moscow: Nauka, 1977.
Wright, S.J., Coordinate Descent Algorithms, Optimizat. Online, 2015 (http://www.optimizationonline. org/DB FILE/2014/12/4679.pdf).
Anikin, A., Dvurechensky, P., Gasnikov, A., et al., Modern Efficient Numerical Approaches to Regularized Regression Problems in Application to Traffic Demands Matrix Calculation From Link Loads, in Proc. Int. Conf. ITAS-2015, Russia, Sochi, September, 2015, arXiv:1508.00858.
Gasnikov, A.V., Dvurechenskii, P.E., and Usmanova, I.N., On Nontriviality of Fast (Accelerated) Randomized Methods, Tr. MFTI, 2016, vol. 8, no. 2, arXiv:1508.02182.
Juditsky, A. and Nemirovski, A., First Order Methods for Nonsmooth Convex Large-scale Optimization. I, II, in Optimization for Machine Learning, Sra, S., Nowozin, S., and Wright, S., Eds., Boston: MIT Press, 2012.
Author information
Authors and Affiliations
Corresponding author
Additional information
Original Russian Text © A.V. Gasnikov, A.A. Lagunovskaya, I.N. Usmanova, F.A. Fedorenko, 2016, published in Avtomatika i Telemekhanika, 2016, No. 10, pp. 57–77.
This paper was recommended for publication by P.S. Shcherbakov, a member of the Editorial Board
Rights and permissions
About this article
Cite this article
Gasnikov, A.V., Lagunovskaya, A.A., Usmanova, I.N. et al. Gradient-free proximal methods with inexact oracle for convex stochastic nonsmooth optimization problems on the simplex. Autom Remote Control 77, 2018–2034 (2016). https://doi.org/10.1134/S0005117916110114
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1134/S0005117916110114