On the sample complexity of uncertain linear and bilinear matrix inequalities

In this paper, we consider uncertain linear and bilinear matrix inequalities which depend in a possibly nonlinear way on a vector of uncertain parameters. Motivated by recent results in statistical learning, we show that probabilistic guaranteed solutions can be obtained by means of randomized algorithms. In particular, we show that the Vapnik-Chevonenkis dimension (VC-dimension) of the two problems is finite, and we compute upper bounds on it. In turn, these bounds allow us to derive explicitly the sample complexity of the problems. Using these bounds, in the second part of the paper, we derive a sequential scheme, based on a sequence of optimization and validation steps. The algorithm is on the same lines of recent schemes proposed for similar problems, but improves both in terms of complexity and generality.

[1]  Mathukumalli Vidyasagar,et al.  Learning and Generalization: With Applications to Neural Networks , 2002 .

[2]  Giuseppe Carlo Calafiore,et al.  The scenario approach to robust control design , 2006, IEEE Transactions on Automatic Control.

[3]  E. Yaz.  Linear Matrix Inequalities In System And Control Theory , 1998, Proceedings of the IEEE.

[4]  Roberto Tempo,et al.  On the Sample Complexity of Probabilistic Analysis and Design Methods , 2010 .

[5]  Dmitry Panchenko,et al.  Improved sample complexity estimates for statistical learning control of uncertain systems , 2000, IEEE Trans. Autom. Control..

[6]  R. Braatz,et al.  A tutorial on linear and bilinear matrix inequalities , 2000 .

[7]  Mario Sznaier,et al.  Randomized Algorithms for Analysis and Control of Uncertain Systems with Applications, Second Edition, Roberto Tempo, Giuseppe Calafiore, Fabrizio Dabbene (Eds.). Springer-Verlag, London (2013), 357, ISBN: 978-1-4471-4609-4 , 2014, Autom..

[8]  Michel Verhaegen,et al.  Robust output-feedback controller design via local BMI optimization , 2004, Autom..

[9]  Michael G. Safonov,et al.  Global optimization for the Biaffine Matrix Inequality problem , 1995, J. Glob. Optim..

[10]  Qing-Guo Wang,et al.  A statistical learning theory approach for uncertain linear and bilinear matrix inequalities , 2013, Autom..

[11]  Roberto Tempo,et al.  Randomized control design through probabilistic validation , 2012, 2012 American Control Conference (ACC).

[12]  Norbert Sauer,et al.  On the Density of Families of Sets , 1972, J. Comb. Theory A.

[13]  Giuseppe Carlo Calafiore,et al.  Research on probabilistic methods for control system design , 2011, Autom..

[14]  Mathukumalli Vidyasagar,et al.  Randomized algorithms for robust controller synthesis using statistical learning theory , 2001, Autom..

[15]  Giuseppe Carlo Calafiore,et al.  Uncertain convex programs: randomized solutions and confidence levels , 2005, Math. Program..

[16]  Vladimir Vapnik,et al.  Statistical learning theory , 1998 .

[17]  Eduardo F. Camacho,et al.  Randomized Strategies for Probabilistic Solutions of Uncertain Feasibility and Optimization Problems , 2009, IEEE Transactions on Automatic Control.

[18]  Vincent D. Blondel,et al.  Probabilistic solutions to some NP-hard matrix problems , 2001, Autom..

[19]  Marco C. Campi,et al.  The Exact Feasibility of Randomized Solutions of Uncertain Convex Programs , 2008, SIAM J. Optim..

[20]  Y. Fujisaki,et al.  Probabilistic robust controller design: probable near minimax value and randomized algorithms , 2003 .