Inductive Conformal Martingales for Change-Point Detection

We consider the problem of quickest change-point detection in data streams. Classical change-point detection procedures, such as CUSUM, Shiryaev-Roberts and Posterior Probability statistics, are optimal only if the change-point model is known, which is an unrealistic assumption in typical applied problems. Instead we propose a new method for change-point detection based on Inductive Conformal Martingales, which requires only the independence and identical distribution of observations. We compare the proposed approach to standard methods, as well as to change-point detection oracles, which model a typical practical situation when we have only imprecise (albeit parametric) information about pre- and post-change data distributions. Results of comparison provide evidence that change-point detection based on Inductive Conformal Martingales is an efficient tool, capable to work under quite general conditions unlike traditional approaches.

[1]  Y. Ritov.  Decision Theoretic Optimality of the Cusum Procedure , 1990 .

[2]  Evgeny Burnaev,et al.  Ensembles of detectors for online detection of transient changes , 2015, International Conference on Machine Vision.

[3]  Evgeny Burnaev,et al.  Model selection for anomaly detection , 2015, International Conference on Machine Vision.

[4]  M. Pollak.  Average Run Lengths of an Optimal Method of Detecting a Change in Distribution. , 1987 .

[5]  Jean-Luc Ville.  Étude critique de la notion de collectif , 1939 .

[6]  Hongjoong Kim,et al.  A novel approach to detection of intrusions in computer networks via adaptive sequential and batch-sequential change-point detection methods , 2006, IEEE Transactions on Signal Processing.

[7]  M. Rosenblatt.  Remarks on Some Nonparametric Estimates of a Density Function , 1956 .

[8]  Michèle Basseville,et al.  Detection of abrupt changes: theory and application , 1993 .

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

[10]  Alexander Gammerman,et al.  Testing Exchangeability On-Line , 2003, ICML.

[11]  S. W. Roberts.  A Comparison of Some Control Chart Procedures , 1966 .

[12]  Evgeny Burnaev,et al.  Influence of resampling on accuracy of imbalanced classification , 2015, International Conference on Machine Vision.

[13]  E. S. Page.  CONTINUOUS INSPECTION SCHEMES , 1954 .

[14]  W. Gasarch,et al.  The Book Review Column 1 Coverage Untyped Systems Simple Types Recursive Types Higher-order Systems General Impression 3 Organization, and Contents of the Book , 2022 .

[15]  Svetha Venkatesh,et al.  Anomaly detection in large-scale data stream networks , 2012, Data Mining and Knowledge Discovery.

[16]  M. Pollak.  Optimal Detection of a Change in Distribution , 1985 .

[17]  G. Lorden.  PROCEDURES FOR REACTING TO A CHANGE IN DISTRIBUTION , 1971 .

[18]  I. Nikiforov,et al.  Sequential Detection of Transient Changes , 2012 .

[19]  Shen-Shyang Ho,et al.  A martingale framework for concept change detection in time-varying data streams , 2005, ICML.

[20]  Alexander Gammerman,et al.  Plug-in martingales for testing exchangeability on-line , 2012, ICML.

[21]  Pedro Casas,et al.  Optimal volume anomaly detection and isolation in large-scale IP networks using coarse-grained measurements , 2010, Comput. Networks.

[22]  Jason L. Speyer,et al.  A generalized Shiryayev sequential probability ratio test for change detection and isolation , 1999, IEEE Trans. Autom. Control..

[23]  Evgeny Burnaev,et al.  Detecting Performance Degradation of Software-Intensive Systems in the Presence of Trends and Long-Range Dependence , 2016, 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW).

[24]  Pavel Erofeev,et al.  Application of Rare Event Anticipation Techniques to Aircraft Health Management , 2014 .

[25]  G. Moustakides.  Optimal stopping times for detecting changes in distributions , 1986 .

[26]  Evgeny Burnaev,et al.  One-Class SVM with Privileged Information and Its Application to Malware Detection , 2016, 2016 IEEE 16th International Conference on Data Mining Workshops (ICDMW).

[27]  A. Shiryaev.  Quickest Detection Problems: Fifty Years Later , 2010 .

[28]  Evgeny Burnaev,et al.  Nonparametric decomposition of quasi-periodic time series for change-point detection , 2015, International Conference on Machine Vision.

[29]  A. Shiryaev.  On Optimum Methods in Quickest Detection Problems , 1963 .