# Change Point Detection in Time Series Data using Autoencoders with a Time-Invariant Representation

arXiv: https://arxiv.org/abs/2008.09524 · HTML: https://mlsanity.com/paper/2008.09524/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Aug 2020.

## Built on by (most cited first, 21 of 21)

- [Automatic Change-Point Detection in Time Series via Deep Learning](https://mlsanity.com/paper/2211.03860/) (2022)
- [Timestamp-supervised Wearable-based Activity Segmentation and Recognition with Contrastive Learning and Order-Preserving Optimal Transport](https://mlsanity.com/paper/2310.09114/) (2023)
- [Memory-free Online Change-point Detection: A Novel Neural Network Approach](https://mlsanity.com/paper/2207.03932/) (2022)
- [PrecTime: A Deep Learning Architecture for Precise Time Series Segmentation in Industrial Manufacturing Operations](https://mlsanity.com/paper/2302.10182/) (2023)
- [Selective Inference for Changepoint detection by Recurrent Neural Network](https://mlsanity.com/paper/2311.14964/) (2023)
- [Latent Neural Stochastic Differential Equations for Change Point Detection](https://mlsanity.com/paper/2208.10317/) (2022)
- [Learning Sinkhorn divergences for supervised change point detection](https://mlsanity.com/paper/2202.04000/) (2022)
- [Self-Supervised Transformer Architecture for Change Detection in Radio Access Networks](https://mlsanity.com/paper/2302.02025/) (2023)
- [Time-series Anomaly Detection based on Difference Subspace between Signal Subspaces](https://mlsanity.com/paper/2303.17802/) (2023)
- [Identifying Stochasticity in Time-Series with Autoencoder-Based Content-aware 2D Representation: Application to Black Hole Data](https://mlsanity.com/paper/2304.11560/) (2023)
- [Cadence: A Practical Time-series Partitioning Algorithm for Unlabeled IoT Sensor Streams](https://mlsanity.com/paper/2112.03360/) (2021)
- [Dynamic Interpretable Change Point Detection](https://mlsanity.com/paper/2211.03991/) (2022)
- [Anomaly Detection for Scalable Task Grouping in Reinforcement Learning-based RAN Optimization](https://mlsanity.com/paper/2312.03277/) (2023)
- [Encoding Binary Events from Continuous Time Series in Rooted Trees using Contrastive Learning](https://mlsanity.com/paper/2401.01242/) (2024)
- [Representation and De-interleaving of Mixtures of Hidden Markov Processes](https://mlsanity.com/paper/2406.00416/) (2024)
- [A Network Science Approach to Granular Time Series Segmentation](https://mlsanity.com/paper/2505.17640/) (2025)
- [Neural Total Variation Distance Estimators for Changepoint Detection in News Data](https://mlsanity.com/paper/2506.18764/) (2025)
- [An Unsupervised Time Series Anomaly Detection Approach for Efficient Online Process Monitoring of Additive Manufacturing](https://mlsanity.com/paper/2510.09977/) (2025)
- [MuRAL-CPD: Active Learning for Multiresolution Change Point Detection](https://mlsanity.com/paper/2601.20686/) (2026)
- [LastAct: Trajectory-Guided Latest-Activity Localization for Real-Time Smart-Home Activity Recognition](https://mlsanity.com/paper/2606.00260/) (2026)
- [High-Dimensional Online Change Point Detection with Adaptive Thresholding and Interpretability](https://mlsanity.com/paper/2609.24278/) (2026)

## Abstract (arXiv)

Change point detection (CPD) aims to locate abrupt property changes in time series data. Recent CPD methods demonstrated the potential of using deep learning techniques, but often lack the ability to identify more subtle changes in the autocorrelation statistics of the signal and suffer from a high false alarm rate. To address these issues, we employ an autoencoder-based methodology with a novel loss function, through which the used autoencoders learn a partially time-invariant representation that is tailored for CPD. The result is a flexible method that allows the user to indicate whether change points should be sought in the time domain, frequency domain or both. Detectable change points include abrupt changes in the slope, mean, variance, autocorrelation function and frequency spectrum. We demonstrate that our proposed method is consistently highly competitive or superior to baseline methods on diverse simulated and real-life benchmark data sets. Finally, we mitigate the issue of false detection alarms through the use of a postprocessing procedure that combines a matched filter and a newly proposed change point score. We show that this combination drastically improves the performance of our method as well as all baseline methods.