Optimal sequential detection in multi-stream data
From MaRDI portal
DOI10.1214/17-AOS1546zbMATH Open1388.62127MaRDI QIDQ134124FDOQ134124
Abstract: Consider a large number of detectors each generating a data stream. The task is to detect online, distribution changes in a small fraction of the data streams. Previous approaches to this problem include the use of mixture likelihood ratios and sum of CUSUMs. We provide here extensions and modifications of these approaches that are optimal in detecting normal mean shifts. We show how the (optimal) detection delay depends on the fraction of data streams undergoing distribution changes as the number of detectors goes to infinity. There are three detection domains. In the first domain for moderately large fractions, immediate detection is possible. In the second domain for smaller fractions, the detection delay grows logarithmically with the number of detectors, with an asymptotic constant extending those in sparse normal mixture detection. In the third domain for even smaller fractions, the detection delay lies in the framework of the classical detection delay formula of Lorden. We show that the optimal detection delay is achieved by the sum of detectability score transformations of either the partial scores or CUSUM scores of the data streams.
Recommendations
- Asymptotically optimal quickest change detection in multistream data -- Part 1: General stochastic models
- Efficient scalable schemes for monitoring a large number of data streams
- Sequential multi-sensor change-point detection
- Optimal detection of changing properties of random sequences. II: Sequential detection
- Sequential detection/isolation of abrupt changes
Cited in
(30)- Large-Scale Datastreams Surveillance via Pattern-Oriented-Sampling
- ocd
- Sequential multi-sensor change-point detection
- Asymptotically optimal quickest change detection in multistream data -- Part 1: General stochastic models
- Isolating changed panels and estimating common change point after sequential detection with FDR control
- Asymptotic optimality theory for active quickest detection with unknown postchange parameters
- scientific article; zbMATH DE number 7370546 (Why is no real title available?)
- False discovery rate approach to dynamic change detection
- A communication-efficient, online changepoint detection method for monitoring distributed sensor networks
- Estimation of common change point and isolation of changed panels after sequential detection
- Efficient scalable schemes for monitoring a large number of data streams
- Breadth-first maximum likelihood sequence detection: Basics
- Common change-point estimation and changed panel isolation after sequential detection in an exponential family
- Sequential Common Change Detection and Isolation of Changed Panels in Panel Data
- Asymptotic statistical properties of communication-efficient quickest detection schemes in sensor networks
- Change detection based on the distribution of \(p\)-values
- Optimal unimodular sequences design method for active sensing systems
- Compound Sequential Change-point Detection in Parallel Data Streams
- Using labeled data to evaluate change detectors in a multivariate streaming environment
- Online multivariate changepoint detection with type I error control and constant time/memory updates per series
- Sequential common change detection, isolation, and estimation in multiple Poisson processes
- Sequential nonparametric tests for a change in distribution: an application to detecting radiological anomalies
- Optimal on-line detection of outside observations
- A combined SR-CUSUM procedure for detecting common changes in panel data
- Quickest changepoint detection in general multistream stochastic models: recent results, applications and future challenges
- Adaptive Partially Observed Sequential Change Detection and Isolation
- Sequential detection of targets in multichannel systems
- Scalable sum-shrinkage schemes for distributed monitoring large-scale data streams
- Bandit Change-Point Detection for Real-Time Monitoring High-Dimensional Data Under Sampling Control
- Robust change detection for large-scale data streams
This page was built for publication: Optimal sequential detection in multi-stream data
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q134124)
