Empirical Variance Minimization with Applications in Variance Reduction and Optimal Control

We study the problem of empirical minimization for variance-type functionals over functional classes. Sharp non-asymptotic bounds for the excess variance are derived under mild conditions. In particular, it is shown that under some restrictions imposed on the functional class fast convergence rates can be achieved including the optimal non-parametric rates for expressive classes in the non-Donsker regime under some additional assumptions. Our main applications include variance reduction and optimal control.

[1]  W. Hoeffding.  The strong law of large numbers for u-statistics. , 1961 .

[2]  Christian P. Robert,et al.  Monte Carlo Statistical Methods , 2005, Springer Texts in Statistics.

[3]  M. Girolami,et al.  Convergence rates for a class of estimators based on Stein’s method , 2016, Bernoulli.

[4]  G. Lugosi,et al.  Robust estimation of U-statistics , 2015, 1504.04580.

[5]  W. Hoeffding.  A Class of Statistics with Asymptotically Normal Distribution , 1948 .

[6]  P. Boyle.  Options: A Monte Carlo approach , 1977 .

[7]  V. Koltchinskii,et al.  Concentration inequalities and asymptotic results for ratio type empirical processes , 2006, math/0606788.

[8]  Nikita Zhivotovskiy,et al.  Variance Reduction in Monte Carlo Estimators via Empirical Variance Minimization , 2018, Doklady Mathematics.

[9]  Stéphan J. Clémençcon.  On U-processes and clustering performance , 2011, NIPS 2011.

[10]  M. Caffarel,et al.  Zero-Variance Principle for Monte Carlo Algorithms , 1999, cond-mat/9911396.

[11]  Denis Belomestny,et al.  Dual Methods for General Optimal Control , 2018 .

[12]  P. Massart,et al.  Concentration inequalities and model selection , 2007 .

[13]  Alan G. White,et al.  The Use of the Control Variate Technique in Option Pricing , 1988, Journal of Financial and Quantitative Analysis.

[14]  W. Wong,et al.  Probability inequalities for likelihood ratios and convergence rates of sieve MLEs , 1995 .

[15]  Shahar Mendelson,et al.  General nonexact oracle inequalities for classes with a subexponential envelope , 2012, 1206.0871.

[16]  O. Bousquet.  A Bennett concentration inequality and its application to suprema of empirical processes , 2002 .

[17]  Nikita Zhivotovskiy.  Optimal learning via local entropies and sample compression , 2017, COLT.

[18]  P. Massart,et al.  Rates of convergence for minimum contrast estimators , 1993 .

[19]  Gábor Lugosi,et al.  Concentration Inequalities - A Nonasymptotic Theory of Independence , 2013, Concentration Inequalities.

[20]  Qiang Liu,et al.  Black-box Importance Sampling , 2016, AISTATS.

[21]  Stéphan Clémençon,et al.  Scaling-up Empirical Risk Minimization: Optimization of Incomplete $U$-statistics , 2015, J. Mach. Learn. Res..

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

[23]  P. Glynn,et al.  Some New Perspectives on the Method of Control Variates , 2002 .

[24]  L. C. G. Rogers,et al.  Pathwise Stochastic Optimal Control , 2007, SIAM J. Control. Optim..

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

[26]  A. Stuart,et al.  Portfolio Selection: Efficient Diversification of Investments , 1959 .

[27]  P. Bartlett,et al.  Local Rademacher complexities , 2005, math/0508275.

[28]  Karthik Sridharan,et al.  Empirical Entropy, Minimax Regret and Minimax Risk , 2013, ArXiv.

[29]  S. Geer.  Empirical Processes in M-Estimation , 2000 .

[30]  Ambuj Tewari,et al.  Smoothness, Low Noise and Fast Rates , 2010, NIPS.

[31]  P. Massart,et al.  Risk bounds for model selection via penalization , 1999 .

[32]  V. Koltchinskii.  Local Rademacher complexities and oracle inequalities in risk minimization , 2006, 0708.0083.

[33]  H. N. Mhaskar,et al.  Neural Networks for Optimal Approximation of Smooth and Analytic Functions , 1996, Neural Computation.

[34]  Wolfgang Hörmann,et al.  Control variates and conditional Monte Carlo for basket and Asian options , 2013 .

[35]  Yuhong Yang,et al.  Information-theoretic determination of minimax rates of convergence , 1999 .

[36]  P. Halmos.  The Theory of Unbiased Estimation , 1946 .

[37]  Sean McKee,et al.  Monte Carlo Methods for Applied Scientists , 2005 .

[38]  G. Lugosi,et al.  Ranking and empirical minimization of U-statistics , 2006, math/0603123.

[39]  Stéphan Clémençon,et al.  A statistical view of clustering performance through the theory of U-processes , 2014, J. Multivar. Anal..

[40]  R. Nickl,et al.  Bracketing Metric Entropy Rates and Empirical Central Limit Theorems for Function Classes of Besov- and Sobolev-Type , 2007 .

[41]  N. Chopin,et al.  Control functionals for Monte Carlo integration , 2014, 1410.2392.

[42]  W. Hoeffding.  Probability Inequalities for sums of Bounded Random Variables , 1963 .

[43]  Antonietta Mira,et al.  Zero variance Markov chain Monte Carlo for Bayesian estimators , 2010, Stat. Comput..

[44]  Johan Segers,et al.  Monte Carlo integration with a growing number of control variates , 2018, Journal of Applied Probability.