Improved sample complexity estimates for statistical learning control of uncertain systems

Probabilistic methods and statistical learning theory have been shown to provide approximate solutions to "difficult" control problems. Unfortunately, the number of samples required in order to guarantee stringent performance levels may be prohibitively large. This paper introduces bootstrap learning methods and the concept of stopping times to drastically reduce the bound on the number of samples required to achieve a performance level. We then apply these results to obtain more efficient algorithms which probabilistically guarantee stability and robustness levels when designing controllers for uncertain systems.

[1]  László Györfi,et al.  A Probabilistic Theory of Pattern Recognition , 1996, Stochastic Modelling and Applied Probability.

[2]  H. Teicher,et al.  Central Limit Theorems , 1997 .

[3]  Robert F. Stengel,et al.  A monte carlo approach to the analysis of control system robustness , 1993, Autom..

[4]  R. Dudley,et al.  Uniform Central Limit Theorems: Notation Index , 2014 .

[5]  Vladimir Vapnik,et al.  Estimation of Dependences Based on Empirical Data: Springer Series in Statistics (Springer Series in Statistics) , 1982 .

[6]  Mathukumalli Vidyasagar,et al.  Statistical learning theory and randomized algorithms for control , 1998 .

[7]  Jon A. Wellner,et al.  Weak Convergence and Empirical Processes: With Applications to Statistics , 1996 .

[8]  R. Tempo,et al.  Probabilistic robustness analysis: explicit bounds for the minimum number of samples , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

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

[10]  B. Ross Barmish,et al.  The uniform distribution: A rigorous justification for its use in robustness analysis , 1996, Math. Control. Signals Syst..

[11]  Vladimir Vapnik,et al.  Chervonenkis: On the uniform convergence of relative frequencies of events to their probabilities , 1971 .

[12]  J. Kuelbs.  Probability on Banach spaces , 1978 .

[13]  Pramod P. Khargonekar,et al.  Randomized algorithms for robust control analysis and synthesis have polynomial complexity , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[14]  V. Koltchinski.  Statistical Learning Control of Uncertain Systems: It is Better Than It Seems , 2000 .

[15]  Kazuoki Azuma.  WEIGHTED SUMS OF CERTAIN DEPENDENT RANDOM VARIABLES , 1967 .

[16]  Mathukumalli Vidyasagar,et al.  A Theory of Learning and Generalization: With Applications to Neural Networks and Control Systems , 1997 .

[17]  A. W. van der Vaart,et al.  Uniform Central Limit Theorems , 2001 .

[18]  H. Chernoff.  A Measure of Asymptotic Efficiency for Tests of a Hypothesis Based on the sum of Observations , 1952 .

[19]  V. V. Jurinskii.  Exponential Bounds for Large Deviations , 1974 .

[20]  D. Panchenko,et al.  Statistical Learning Control of Uncertain Systems : It is Better Than It Seems , 1999 .

[21]  Minyue Fu,et al.  Worst-case properties of the uniform distribution and randomized algorithms for robustness analysis , 1998, Math. Control. Signals Syst..

[22]  J. Tsitsiklis,et al.  NP-hardness of some linear control design problems , 1995, Proceedings of 1995 34th IEEE Conference on Decision and Control.

[23]  J. Tsitsiklis,et al.  Three problems on the decidability and complexity of stability , 1999 .

[24]  Chaouki T. Abdallah,et al.  Statistical controller design for the linear benchmark problem , 1999, Proceedings of the 38th IEEE Conference on Decision and Control (Cat. No.99CH36304).

[25]  Vladimir Koltchinskii,et al.  Rademacher penalties and structural risk minimization , 2001, IEEE Trans. Inf. Theory.

[26]  Kameshwar Poolla,et al.  Statistical validation for uncertainty models , 1995 .

[27]  Kemin Zhou,et al.  A probabilistic approach to robust control , 1997, Proceedings of the 36th IEEE Conference on Decision and Control.

[28]  John N. Tsitsiklis,et al.  A survey of computational complexity results in systems and control , 2000, Autom..

[29]  R. Dudley.  Central Limit Theorems for Empirical Measures , 1978 .

[30]  O. Toker,et al.  On the NP-hardness of solving bilinear matrix inequalities and simultaneous stabilization with static output feedback , 1995, Proceedings of 1995 American Control Conference - ACC'95.