Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

A Contrastive Approach to Online Change Point Detection

About

We suggest a novel procedure for online change point detection. Our approach expands an idea of maximizing a discrepancy measure between points from pre-change and post-change distributions. This leads to a flexible procedure suitable for both parametric and nonparametric scenarios. We prove non-asymptotic bounds on the average running length of the procedure and its expected detection delay. The efficiency of the algorithm is illustrated with numerical experiments on synthetic and real-world data sets.

Artur Goldman, Nikita Puchkin, Valeriia Shcherbakova, Uliana Vinogradova• 2022

Related benchmarks

TaskDatasetResultRank
Change Point DetectionSynthetic data Example 1 (test)
FA Rate0.00e+0
5
Change Point DetectionSynthetic data Example 2 (test)
FA0.00e+0
5
Change Point DetectionSynthetic data Example 3 (test)
FA Rate0.00e+0
5
Change Point DetectionSynthetic data Example 4 (test)
False Alarm Rate0.00e+0
5
Change Point DetectionRoom Occupancy (test)
False Alarm Rate1
5
Change Point DetectionCENSREC-1-C Clean Record
FA Rate0.00e+0
5
Change Point DetectionCENSREC-1-C SNR 20
FA0.00e+0
5
Change Point DetectionCENSREC-1-C SNR 15
FA0.00e+0
5
Human activity detectionWISDM
FA Rate1
5
Showing 9 of 9 rows

Other info

Follow for update