Adaptive online kernel changepoint detection
Abstract
We propose an adaptive online kernel-based changepoint detection method for streaming data that is capable of detecting a broad range of changes in the underlying data distribution. The method maintains a recursively-weighted reproducing kernel Hilbert space representation of observations and adaptively updates the forgetting factor through a gradient-based procedure driven by a maximum mean discrepancy-type statistic between the current observation and the weighted empirical distribution of the past. This self-tuning mechanism allows the detector to adapt its effective memory and responsiveness to changes in the underlying process. Simulation results demonstrate that the proposed method achieves strong detection performance across a wide range of distributional changes. Further, our proposed approach maintains constant computational and storage cost through recursive updates, and is very computationally efficient in comparison to competing methods. Experiments on both simulated data and benchmark real-world datasets show improved performance over several other leading kernel-based methods.
Related
- Bagel: A Fast Bayesian Online Changepoint Detection Algorithm for Linear Models via Posterior Merging
- Co-Evolving Zero-Day Jamming: Adaptive Attack Synthesis and Graph Attention-Based Online Detection
- Changepoint detection methods as a practical tool for assessment of ecological restoration
- FOCUS: Fixed-Confidence Online Causal Learning Using Sequential Adaptive Interventions
- Transformers as Adaptive Kernel Learners across Tasks for In-Context Learning
BibTeX
@misc{jiang2026adaptive,
title = {{Adaptive online kernel changepoint detection}},
author = {Qianqian Jiang and Dean Bodenham},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.22545},
eprint = {2609.22545},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.22545},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Adaptive online kernel changepoint detection
AU - Qianqian Jiang
AU - Dean Bodenham
PY - 2026
DA - 2026/09/18
PB - arXiv
DO - 10.48550/arxiv.2609.22545
UR - https://arxiv.org/abs/2609.22545
AB - We propose an adaptive online kernel-based changepoint detection method for streaming data that is capable of detecting a broad range of changes in the underlying data distribution. The method maintains a recursively-weighted reproducing kernel Hilbert space representation of observations and adaptively updates the forgetting factor through a gradient-based procedure driven by a maximum mean discrepancy-type statistic between the current observation and the weighted empirical distribution of the past. This self-tuning mechanism allows the detector to adapt its effective memory and responsiveness to changes in the underlying process. Simulation results demonstrate that the proposed method achieves strong detection performance across a wide range of distributional changes. Further, our proposed approach maintains constant computational and storage cost through recursive updates, and is very computationally efficient in comparison to competing methods. Experiments on both simulated data and benchmark real-world datasets show improved performance over several other leading kernel-based methods.
ER -