Robust Sequential Change-Point Detection by Convex Optimization
arXiv:1701.06952 · doi:10.3390/e20020108
Abstract
We address the computational challenge of finding the robust sequential change-point detection procedures when the pre- and post-change distributions are not completely specified. Earlier works [veeravalli 1994] and [Unnikrishnan 2011] establish the general conditions for robust procedures which include finding a pair of least favorable distributions (LFDs). However, in the multi-dimensional setting, it is hard to find such LFDs computationally. We present a method based on convex optimization that addresses this issue when the distributions are Gaussian with unknown parameters from pre-specified uncertainty sets. We also establish theoretical properties of our robust procedures, and numerical examples demonstrate their good performance.
arXiv commentAccepted by ISIT 2017
References in corpus (10)
- Sequential multi-sensor change-point detection
- Detecting simultaneous variant intervals in aligned sequences
- Optimal Joint Target Detection and Parameter Estimation By MIMO Radar
- Minimax Robust Quickest Change Detection
- Online Learning for Time Series Prediction
- Sequential change-point detection when unknown parameters are present in the pre-change distribution
- Nonanticipating estimation applied to sequential analysis and changepoint detection
- Sequential anomaly detection in the presence of noise and limited feedback
- Large-Scale Multi-Stream Quickest Change Detection via Shrinkage Post-Change Estimation
- Change Detection via Affine and Quadratic Detectors
Cited by in corpus (4)
- Real-Time Nonparametric Anomaly Detection in High-Dimensional Settings
- Quickest Change Detection for Multiple Data Streams Using the James-Stein Estimator
- Predictive change point detection for heterogeneous data
- Signal Processing Challenges and Examples for {\it in-situ} Transmission Electron Microscopy