Online Neural Networks for Change-Point Detection
arXiv:2010.01388 · doi:10.1007/s10994-026-07000-6
Abstract
Moments when a time series changes its behavior are called change points. Occurrence of change point implies that the state of the system is altered and its timely detection might help to prevent unwanted consequences. In this paper, we present two change-point detection approaches based on neural networks and online learning. These algorithms demonstrate linear computational complexity and are suitable for change-point detection in large time series. We compare them with the best known algorithms on various synthetic and real world data sets. Experiments show that the proposed methods outperform known approaches. We also prove the convergence of the algorithms to the optimal solutions and describe conditions rendering current approach more powerful than offline one.
This version of the article has been submitted to the journal but is not the Version of Record and does not reflect peer-review improvements, post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/s10994-026-07000-6
References in corpus (16)
- Natural Scales in Geographical Patterns
- Optimal detection of changepoints with a linear computational cost
- Selective review of offline change point detection methods
- Searching for Exotic Particles in High-Energy Physics with Deep Learning
- Wild binary segmentation for multiple change-point detection
- Change-Point Detection in Time-Series Data by Relative Density-Ratio Estimation
- The High Time Resolution Universe Pulsar Survey I: System configuration and initial discoveries
- DCdetector: Dual Attention Contrastive Representation Learning for Time Series Anomaly Detection
- Fifty Years of Pulsar Candidate Selection: From simple filters to a new principled real-time classification approach
- Self-supervised Contrastive Representation Learning for Semi-supervised Time-Series Classification
- Time Series Change Point Detection with Self-Supervised Contrastive Predictive Coding
- Change Point Detection in Time Series Data using Autoencoders with a Time-Invariant Representation
- Deep density ratio estimation for change point detection
- InDiD: Instant Disorder Detection via Representation Learning
- Latent Neural Stochastic Differential Equations for Change Point Detection
- Generalization of Change-Point Detection in Time Series Data Based on Direct Density Ratio Estimation