Uncertain convex programs: randomized solutions and confidence levels
 Giuseppe Calafiore,
 M.C. Campi
 … show all 2 hide
Purchase on Springer.com
$39.95 / €34.95 / £29.95*
Rent the article at a discount
Rent now* Final gross prices may vary according to local VAT.
Abstract.
Many engineering problems can be cast as optimization problems subject to convex constraints that are parameterized by an uncertainty or ‘instance’ parameter. Two main approaches are generally available to tackle constrained optimization problems in presence of uncertainty: robust optimization and chanceconstrained optimization. Robust optimization is a deterministic paradigm where one seeks a solution which simultaneously satisfies all possible constraint instances. In chanceconstrained optimization a probability distribution is instead assumed on the uncertain parameters, and the constraints are enforced up to a prespecified level of probability. Unfortunately however, both approaches lead to computationally intractable problem formulations.
In this paper, we consider an alternative ‘randomized’ or ‘scenario’ approach for dealing with uncertainty in optimization, based on constraint sampling. In particular, we study the constrained optimization problem resulting by taking into account only a finite set of N constraints, chosen at random among the possible constraint instances of the uncertain problem. We show that the resulting randomized solution fails to satisfy only a small portion of the original constraints, provided that a sufficient number of samples is drawn. Our key result is to provide an efficient and explicit bound on the measure (probability or volume) of the original constraints that are possibly violated by the randomized solution. This volume rapidly decreases to zero as N is increased.
 Apkarian, P., Tuan, H.D.: Parameterized LMIs in control theory. SIAM J. Control Optim 38 (4), 1241–1264 (2000) CrossRef
 Barmish, B.R., Sherbakov, P.: On avoiding vertexization of robustness problems: the approximate feasibility concept. IEEE Trans. Aut. Control 47, 819–824 (2002) CrossRef
 BenTal, A., Nemirovski, A.: Robust truss topology design via semidefinite programming. SIAM J. Optim 7 (4), 991–1016 (1997) CrossRef
 BenTal, A., Nemirovski, A.: Robust convex optimization. Math. Oper. Res. 23 (4), 769–805 (1998)
 BenTal, A., Nemirovski, A.: Robust solutions of uncertain linear programs. Oper. Res. Lett. 25 (1), 1–13 (1999) CrossRef
 BenTal, A., Nemirovski, A.: On tractable approximations of uncertain linear matrix inequalities affected by interval uncertainty. SIAM J. Optim. 12 (3), 811–833 (2002) CrossRef
 Calafiore, G., Campi, M.C., El Ghaoui, L.: Identification of reliable predictor models for unknown systems: a dataconsistency approach based on learning theory. In: Proceedings of IFAC 2002 World Congress, Barcelona, Spain, 2002
 Calafiore, G., Dabbene, F.: A probabilistic framework for problems with real structured uncertainty in systems and control. Automatica 38 (8), 1265–1276 (2002) CrossRef
 Calafiore, G., Polyak, B.: Stochastic algorithms for exact and approximate feasibility of robust LMIs. IEEE Trans. Aut. Control 46 (11), 1755–1759, November 2001 CrossRef
 Charnes, A., Cooper, W.W.: Chance constrained programming. Manag. Sci. 6, 73–79 (1959) CrossRef
 Dantzig, G.B.: Linear programming under uncertainty. Manage. Sci. 1, 197–206 (1955)
 de Farias, D.P., Van Roy, B.: On constraint sampling in the linear programming approach to approximate dynamic programming. Technical Report Dept. Management Sci. Stanford University, 2001
 El Ghaoui, L., Calafiore, G.: Robust filtering for discretetime systems with bounded noise and parametric uncertainty IEEE Trans. Aut. Control 46 (7), 1084–1089, July 2001 CrossRef
 El Ghaoui, L., Lebret, H.: Robust solutions to leastsquares problems with uncertain data. SIAM J. Matrix Anal. Appl. 18 (4), 1035–1064 (1997) CrossRef
 El Ghaoui, L., Lebret, H.: Robust solutions to uncertain semidefinite programs. SIAM J. Optim. 9 (1), 33–52 (1998) CrossRef
 Goberna, M.A., Lopez, M.A.: Linear SemiInfinite Optimization. Wiley, 1998
 Goemans, M.X., Williamson, D.P.: .878approximation for MAX CUT and MAX 2SAT. Proc. 26th ACM Sym. Theor. Computing, 1994, pp. 422–431
 Guestrin, C., Koller, D., Parr, R.: Efficient solution algorithms for factored MDPs. To appear in J. Artificial Intelligence Research, submitted 2002
 Hoeffding, W.: Probability inequalities for sums of bounded random variables. J. Am. Statistical Association 58, 13–30 (1963)
 Kelley, J.E.: The cuttingplane method for solving convex programs. J. Soc. Ind. Appl. Math. 8 (4), 703–712 (1961)
 Krishnan, K.: Linear programming (LP) approaches to semidefinite programming (SDP) problems. Rensselaer Polytechnic Institute, Ph.D. Thesis, Troy, New York, 2001
 Lobo, M., Vandenberghe, L., Boyd, S., Lebret, H.: Applications of secondorder cone programming. Linear Algebra and its Applications 284, 193–228 November 1998
 Luebbecke, M.E., Desrosiers, J.: Selected topics in column generation. Les Cahiers de GERAD G200264, 2002, pp. 193–228
 Morrison, J.R., Kumar, P.R.: New linear program performance bound for queuing networks J. Optim. Theor. Appl. 100 (3), 575–597 (1999) CrossRef
 Motwani, R., Raghavan, P.: Randomized Algorithms. Cambridge University Press, Cambridge, 1995
 Mulvey, J.M., Vanderbei, R.J., Zenios, S.A.: Robust optimization of largescale systems. Oper. Res. 43, 264–281 (1995)
 Prékopa, A.: Stochastic Programming. Kluwer, 1995
 Schuurmans, D., Patrascu, R.: Direct valueapproximation for factored MDPs. Adv. Neural Infor. Processing (NIPS), 2001
 Shapiro, A., HomemdeMello, T.: On rate of convergence of Monte Carlo approximations of stochastic programs. SIAM J. Optim. 11, 70–86 (2000) CrossRef
 Still, G.: Discretization in semiinfinite programming: Rate of convergence. Math. Program. Ser. A 91, 53–69 (2001)
 Todd, M.J.: Semidefinite optimization. Acta Numerica 10, 515–560 (2001) CrossRef
 Vandenberghe, L., Boyd, S.: Semidefinite programming. SIAM Rev. 38 (1), 49–95, March 1996
 Vapnik, V.N., Ya.Chervonenkis, A.: On the uniform convergence of relative frequencies to their probabilities. Theor. Probab. Appl. 16 (2), 264–280 (1971)
 Vidyasagar, M.: A theory of learning and generalization. SpringerVerlag, London, 1997
 Title
 Uncertain convex programs: randomized solutions and confidence levels
 Journal

Mathematical Programming
Volume 102, Issue 1 , pp 2546
 Cover Date
 20050101
 DOI
 10.1007/s101070030499y
 Print ISSN
 00255610
 Online ISSN
 14364646
 Publisher
 SpringerVerlag
 Additional Links
 Topics
 Industry Sectors
 Authors

 Giuseppe Calafiore ^{(1)}
 M.C. Campi ^{(2)}
 Author Affiliations

 1. Dipartimento di Automatica e Informatica, Politecnico di Torino, corso Duca degli Abruzzi, 24, 10129, Torino, Italy
 2. Dipartimento di Automatica per l’Automazione, Università di Brescia, via Branze 38, 25123, Brescia, Italy