Minimax optimal sequential hypothesis tests for Markov processes

Under mild Markov assumptions, sufficient conditions for strict minimax optimality of sequential tests for multiple hypotheses under distributional uncertainty are derived. First, the design of optimal sequential tests for simple hypotheses is revisited and it is shown that the partial derivatives of the corresponding cost function are closely related to the performance metrics of the underlying sequential test. Second, an implicit characterization of the least favorable distributions for a given testing policy is stated. By combining the results on optimal sequential tests and least favorable distributions, sufficient conditions for a sequential test to be minimax optimal under general distributional uncertainties are obtained. The cost function of the minimax optimal test is further identified as a generalized $f$-dissimilarity and the least favorable distributions as those that are most similar with respect to this dissimilarity. Numerical examples for minimax optimal sequential tests under different uncertainties illustrate the theoretical results.

[1]  Michael Muma,et al.  Robust Statistics for Signal Processing , 2018 .

[2]  H. Vincent Poor,et al.  On the Equivalence of $f$-Divergence Balls and Density Bands in Robust Detection , 2018, 2018 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[3]  R. Kunsch.  High-Dimensional Function Approximation: Breaking the Curse with Monte Carlo Methods , 2017, 1704.08213.

[4]  Abdelhak M. Zoubir,et al.  Minimax Robust Sequential Hypothesis Testing Under Density Band Uncertainties , 2017 .

[5]  A. Kharin.  On Robustifying of the Sequential Probability Ratio Test for a Discrete Model under "Contaminations" , 2016 .

[6]  I. Csiszár,et al.  MEASURING DISTRIBUTION MODEL RISK , 2016 .

[7]  Hugo Van hamme,et al.  Noise robust exemplar matching with alpha-beta divergence , 2016, Speech Commun..

[8]  Abdelhak M. Zoubir,et al.  Old Bands, New Tracks—Revisiting the Band Model for Robust Hypothesis Testing , 2015, IEEE Transactions on Signal Processing.

[9]  E. Ronchetti,et al.  Robust statistics: a selective overview and new directions , 2015 .

[10]  Ilias S. Kotsireas,et al.  Interdisciplinary Topics in Applied Mathematics, Modeling and Computational Science , 2015 .

[11]  Abdelhak M. Zoubir,et al.  A Linear Programming Approach to Sequential Hypothesis Testing , 2015, 1501.05870.

[12]  Taposh Banerjee,et al.  Data-Efficient Minimax Quickest Change Detection With Composite Post-Change Distribution , 2015, IEEE Transactions on Information Theory.

[13]  M. Basseville,et al.  Sequential Analysis: Hypothesis Testing and Changepoint Detection , 2014 .

[14]  Christian H. Weiß,et al.  Parameter estimation for binomial AR(1) models with applications in finance and industry , 2013 .

[15]  A. Shiryaev,et al.  The optimal decision rule in the Kiefer–Weiss problem for a Brownian motion , 2013 .

[16]  Michael Muma,et al.  Robust Estimation in Signal Processing: A Tutorial-Style Treatment of Fundamental Concepts , 2012, IEEE Signal Processing Magazine.

[17]  A. Tartakovsky,et al.  Almost optimal sequential tests of discrete composite hypotheses , 2012, 1204.5291.

[18]  A. Tartakovsky,et al.  Nearly Minimax One-Sided Mixture-Based Sequential Tests , 2011, 1110.0902.

[19]  Kush R. Varshney,et al.  Bayes Risk Error is a Bregman Divergence , 2011, IEEE Transactions on Signal Processing.

[20]  Sean P. Meyn,et al.  Minimax Robust Quickest Change Detection , 2009, IEEE Transactions on Information Theory.

[21]  Mark D. Reid,et al.  Information, Divergence and Risk for Binary Experiments , 2009, J. Mach. Learn. Res..

[22]  Nguyen Huy Chieu.  The Fréchet and limiting subdifferentials of integral functionals on the spaces L1 (Ω, E) , 2009 .

[23]  Andrey Novikov.  Optimal Sequential Multiple Hypothesis Tests , 2009, Kybernetika.

[24]  B. E. Brodsky,et al.  Minimax Sequential Tests for Many Composite Hypotheses. II , 2008 .

[25]  B. Brodsky,et al.  Minimax Methods for Multihypothesis Sequential Testing and Change-Point Detection Problems , 2008 .

[26]  V. Yohai,et al.  Robust Statistics: Theory and Methods , 2006 .

[27]  Martin J. Wainwright,et al.  ON surrogate loss functions and f-divergences , 2005, math/0510521.

[28]  L. Pardo.  Statistical Inference Based on Divergence Measures , 2005 .

[29]  Jiri Matas,et al.  WaldBoost - learning for time constrained sequential detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[30]  Robert F. Stengel,et al.  Smooth function approximation using neural networks , 2005, IEEE Transactions on Neural Networks.

[31]  A. H. Siddiqi.  Elements of Metric Spaces , 2003 .

[32]  X. Rong Li,et al.  Sequential detection of targets in multichannel systems , 2003, IEEE Trans. Inf. Theory.

[33]  H. Vos.  A minimax procedure in the context of sequential testing problems in psychodiagnostics. , 2001, The British journal of mathematical and statistical psychology.

[34]  M. C. Jones,et al.  Robust and efficient estimation by minimising a density power divergence , 1998 .

[35]  Nikolaos S. Papageorgiou.  Convex integral functionals , 1997 .

[36]  Peter M. Clarkson,et al.  A robust sequential detection algorithm for cardiac arrhythmia classification , 1996, IEEE Transactions on Biomedical Engineering.

[37]  B. K. Ghosh,et al.  Handbook of sequential analysis , 1991 .

[38]  I. Pavlov.  Sequential Procedure of Testing Composite Hypotheses with Applications to the Kiefer–Weiss Problem , 1991 .

[39]  V. P. Dragalin,et al.  Asymptotic Solution of the Kiefer–Weiss Problem for Processes with Independent Increments , 1988 .

[40]  Ludwik Kurz,et al.  A robust approach to sequential detection , 1988, IEEE Trans. Acoust. Speech Signal Process..

[41]  Norbert Schmitz.  Minimax sequential tests of composite hypotheses on the drift of a Wiener process , 1987 .

[42]  S.A. Kassam,et al.  Robust techniques for signal processing: A survey , 1985, Proceedings of the IEEE.

[43]  Saleem A. Kassam,et al.  Robust hypothesis testing for bounded classes of probability densities , 1981, IEEE Trans. Inf. Theory.

[44]  H. Vincent Poor,et al.  Robust decision design using a distance criterion , 1980, IEEE Trans. Inf. Theory.

[45]  V. David VandeLinde,et al.  Robust sequential detection of signals in noise , 1979, IEEE Trans. Inf. Theory.

[46]  T. Nemetz,et al.  f-dissimilarity: A generalization of the affinity of several distributions , 1978 .

[47]  F. Österreicher,et al.  On the construction of least favourable pairs of distributions , 1978 .

[48]  G. Lorden.  2-SPRT'S and The Modified Kiefer-Weiss Problem of Minimizing an Expected Sample Size , 1976 .

[49]  P. J. Huber,et al.  Minimax Tests and the Neyman-Pearson Lemma for Capacities , 1973 .

[50]  Bevan K. Youse,et al.  Introduction to real analysis , 1972 .

[51]  J. Andel.  Sequential Analysis , 2022, The SAGE Encyclopedia of Research Design.

[52]  P. J. Huber.  A Robust Version of the Probability Ratio Test , 1965 .

[53]  W. Rudin.  Principles of mathematical analysis , 1964 .

[54]  M. H. DeGroot.  Minimax Sequential Tests of Some Composite Hypotheses , 1960 .

[55]  N. L. Johnson,et al.  A Minimax‐Regret Procedure for Choosing between Two Populations Using Sequential Sampling , 1957 .

[56]  J. Kiefer,et al.  Some Properties of Generalized Sequential Probability Ratio Tests , 1957 .

[57]  J. Kiefer,et al.  Sequential Decision Problems for Processes with Continuous Time Parameter Problems of Estimation , 1953 .

[58]  J. Kiefer,et al.  Sequential Decision Problems for Processes with Continuous time Parameter. Testing Hypotheses , 1953 .