Abstract
We study a first-order primal-dual subgradient method to optimize risk-constrained risk-penalized optimization problems, where risk is modeled via the popular conditional value at risk (CVaR) measure. The algorithm processes independent and identically distributed samples from the underlying uncertainty in an online fashion and produces an \(\eta /\sqrt{K}\)-approximately feasible and \(\eta /\sqrt{K}\)-approximately optimal point within K iterations with constant step-size, where \(\eta \) increases with tunable risk-parameters of CVaR. We find optimized step sizes using our bounds and precisely characterize the computational cost of risk aversion as revealed by the growth in \(\eta \). Our proposed algorithm makes a simple modification to a typical primal-dual stochastic subgradient algorithm. With this mild change, our analysis surprisingly obviates the need to impose a priori bounds or complex adaptive bounding schemes for dual variables to execute the algorithm as assumed in many prior works. We also draw interesting parallels in sample complexity with that for chance-constrained programs derived in the literature with a very different solution architecture.
Similar content being viewed by others
Notes
requires that we sample \(\omega \) once more for the z-update.
The integrality of K is ignored for notational convenience.
Lemma 2.1 provides sufficient conditions for the existence of such a saddle point.
\(\mathrm{CVaR}\) of any random variable can only vary between the mean and the maximum value that random variable can take.
References
Ahmadi-Javid, A.: Entropic value-at-risk: a new coherent risk measure. J. Optim. Theory Appl. 155(3), 1105–1123 (2012)
Baes, M., Bürgisser, M., Nemirovski, A.: A randomized mirror-prox method for solving structured large-scale matrix saddle-point problems. SIAM J. Optim. 23(2), 934–962 (2013)
Bedi, A.S., Koppel, A., Rajawat, K.: Nonparametric compositional stochastic optimization. arXiv preprint arXiv:1902.06011 (2019)
Ben-Tal, A., El Ghaoui, L., Nemirovski, A.: Robust Optimization, vol. 28. Princeton University Press, Princeton (2009)
Bertsekas, D.P.: Stochastic optimization problems with nondifferentiable cost functionals. J. Optim. Theory Appl. 12(2), 218–231 (1973)
Bonnans, J.F., Shapiro, A.: Perturbation Analysis of Optimization Problems. Springer, Berlin (2013)
Boob, D., Deng, Q., Lan, G.: Stochastic first-order methods for convex and nonconvex functional constrained optimization. arXiv preprint arXiv:1908.02734 (2019)
Borkar, V.S., Meyn, S.P.: The ode method for convergence of stochastic approximation and reinforcement learning. SIAM J. Control Optim. 38(2), 447–469 (2000)
Boyd, S., Mutapcic, A.: Subgradient methods. Lecture notes of EE364b, Stanford University, Winter Quarter 2007 (2006)
Calafiore, G., Campi, M.C.: Uncertain convex programs: randomized solutions and confidence levels. Math. Program. 102(1), 25–46 (2005)
Campi, M.C., Garatti, S.: The exact feasibility of randomized solutions of uncertain convex programs. SIAM J. Optim. 19(3), 1211–1230 (2008)
Charnes, A., Cooper, W.W.: Chance-constrained programming. Manag. Sci. 6(1), 73–79 (1959)
Doan, T.T., Bose, S., Nguyen, D.H., Beck, C.L.: Convergence of the iterates in mirror descent methods. IEEE Control Syst. Lett. 3(1), 114–119 (2018)
Dominguez-Garcia, A.D., Hadjicostis, C.N.: Distributed matrix scaling and application to average consensus in directed graphs. IEEE Trans. Autom. Control 58(3), 667–681 (2013)
Ermoliev, Y.M.: Methods of stochastic programming (1976)
Hadjiyiannis, M.J., Goulart, P.J., Kuhn, D.: An efficient method to estimate the suboptimality of affine controllers. IEEE Trans. Autom. Control 56(12), 2841–2853 (2011)
Hanasusanto, G.A., Kuhn, D., Wiesemann, W.: A comment on “computational complexity of stochastic programming problems”. Math. Program. 159(1–2), 557–569 (2016)
Hiriart-Urruty, J.B., Lemaréchal, C.: Convex Analysis and Minimization Algorithms I: Fundamentals, vol. 305. Springer, Berlin (2013)
Johnson, R., Zhang, T.: Accelerating stochastic gradient descent using predictive variance reduction. In: Advances in Neural Information Processing Systems, pp. 315–323 (2013)
Kalogerias, D.S., Powell, W.B.: Recursive optimization of convex risk measures: mean-semideviation models. arXiv preprint arXiv:1804.00636 (2018)
Kiefer, J., Wolfowitz, J., et al.: Stochastic estimation of the maximum of a regression function. Ann. Math. Stat. 23(3), 462–466 (1952)
Kisiala, J.: Conditional value-at-risk: theory and applications. arXiv preprint arXiv:1511.00140 (2015)
Koppel, A., Sadler, B.M., Ribeiro, A.: Proximity without consensus in online multiagent optimization. IEEE Trans. Signal Process. 65(12), 3062–3077 (2017)
Kushner, H., Yin, G.G.: Stochastic Approximation and Recursive Algorithms and Applications, vol. 35. Springer, Berlin (2003)
Mafusalov, A., Uryasev, S.: Buffered probability of exceedance: mathematical properties and optimization. SIAM J. Optim. 28(2), 1077–1103 (2018)
Mahdavi, M., Jin, R., Yang, T.: Trading regret for efficiency: online convex optimization with long term constraints. J. Mach. Learn. Res. 13(1), 2503–2528 (2012)
Miller, C.W., Yang, I.: Optimal control of conditional value-at-risk in continuous time. SIAM J. Control. Optim. 55(2), 856–884 (2017)
Nedić, A., Lee, S.: On stochastic subgradient mirror-descent algorithm with weighted averaging. SIAM J. Optim. 24(1), 84–107 (2014)
Nedić, A., Ozdaglar, A.: Distributed subgradient methods for multi-agent optimization. IEEE Trans. Autom. Control 54(1), 48–61 (2009)
Nedić, A., Ozdaglar, A.: Subgradient methods for saddle-point problems. J. Optim. Theory Appl. 142(1), 205–228 (2009)
Nemirovski, A., Juditsky, A., Lan, G., Shapiro, A.: Robust stochastic approximation approach to stochastic programming. SIAM J. Optim. 19(4), 1574–1609 (2009)
Nesterov, Y.: Introductory Lectures on Convex Optimization: A Basic Course. Springer (2004)
Ogryczak, W., Ruszczyński, A.: From stochastic dominance to mean-risk models: semideviations as risk measures. Eur. J. Oper. Res. 116(1), 33–50 (1999)
Robbins, H., Monro, S.: A stochastic approximation method. Ann. Math. Stat. 22, 400–407 (1951)
Robbins, H., Siegmund, D.: A convergence theorem for non negative almost supermartingales and some applications. In: Optimizing Methods in Statistics, pp. 233–257. Elsevier (1971)
Rockafellar, R.T., Uryasev, S.: Conditional value-at-risk for general loss distributions. J. Bank. Finance 26(7), 1443–1471 (2002)
Ruszczyński, A., Shapiro, A.: Optimization of convex risk functions. Math. Oper. Res. 31(3), 433–452 (2006)
Schmidt, M., Le Roux, N., Bach, F.: Minimizing finite sums with the stochastic average gradient. Math. Program. 162(1–2), 83–112 (2017)
Shapiro, A., Philpott, A.: A tutorial on stochastic programming. Manuscript. Available at www2.isye.gatech.edu/ashapiro/publications.html17 (2007)
Skaf, J., Boyd, S.P.: Design of affine controllers via convex optimization. IEEE Trans. Autom. Control 55(11), 2476–2487 (2010)
Sun, T., Sun, Y., Yin, W.: On Markov chain gradient descent. In: Advances in Neural Information Processing Systems, pp. 9896–9905 (2018)
Xu, Y.: Primal-dual stochastic gradient method for convex programs with many functional constraints. arXiv preprint arXiv:1802.02724v1 (2018)
Yamashita, S., Hatanaka, T., Yamauchi, J., Fujita, M.: Passivity-based generalization of primal-dual dynamics for non-strictly convex cost functions. Automatica 112, 108712 (2020)
Yu, H., Neely, M., Wei, X.: Online convex optimization with stochastic constraints. In: Advances in Neural Information Processing Systems, pp. 1428–1438 (2017)
Zhang, T., Uryasev, S., Guan, Y.: Derivatives and subderivatives of buffered probability of exceedance. Oper. Res. Lett. 47(2), 130–132 (2019)
Zinkevich, M.: Online convex programming and generalized infinitesimal gradient ascent. In: Proceedings of the 20th International Conference on Machine Learning (ICML-03), pp. 928–936 (2003)
Acknowledgements
We thank Eilyan Bitar, Rayadurgam Srikant, Tamer Başar, and Stan Uryasev for helpful discussions. This work was partially supported by the National Science Foundation under grant no. CAREER-2048065, the International Institute of Carbon-Neutral Energy Research (\(\hbox {I}^2\)CNER), and the Power System Engineering Research Center (PSERC).
Author information
Authors and Affiliations
Corresponding author
Additional information
Communicated by Xiaolu Tan.
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
Madavan, A.N., Bose, S. A Stochastic Primal-Dual Method for Optimization with Conditional Value at Risk Constraints. J Optim Theory Appl 190, 428–460 (2021). https://doi.org/10.1007/s10957-021-01888-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10957-021-01888-x