With the growing prevalence of sensors and other devices collecting data in real-time, automated data analysis methods with theoretically justified performance guarantees are in constant demand. Often a key question with such streaming data is whether they show evidence of anomalous behaviour. This could, e.g., be due to malicious bot activity on a website; early warning of potential equipment failure or detection of methane leakages. In such cases, These and other motivating examples share a common feature which is not accommodated by classical point anomaly models in statistics: the anomaly may not simply be an 'outlying' observation, but rather a distinctive pattern observed over consecutive observations. 
 
The strategic vision for this programme grant is to establish the statistical foundations for Detecting Anomalous Structure in Streaming data settings (DASS). The DASS programme brings together researchers from four of the UK's leading universities in statistics: Lancaster University, the London School of Economics and Political Science, the University of Bristol and the University of Warwick. This £4M initiative, funded by the Engineering and Physical Sciences Research Council (EPSRC), the four participating institutions, and several dedicated project partners, will run for five years (2024-2029).
 
A key element of DASS's approach is the active engagement of numerous partners, for whom the detection and interpretation of anomalous structures are vital. These partners represent sectors including Computing and Communication Technology, Energy and Environment, and Security.

​

Related Background​

​

Change-point detection for graphical models in the presence of missing values

Malte Londschien, Solt Kovács, Peter Bühlmann (2021) Journal of Computational and Graphical Statistics, 30(3), 768–779.

High-dimensional changepoint estimation with heterogeneous missingness

Bertille Follain, Tengyao Wang, Richard J. Samworth (2022) Journal of the Royal Statistical Society Series B: Statistical Methodology, 84(3), 1023–1055

High dimensional change point estimation via sparse projection

Tengyao Wang, Richard J. Samworth (2018) Journal of the Royal Statistical Society Series B: Statistical Methodology, 80(1), 57–83

Scalable change-point and anomaly detection in cross-correlated data with an application to condition monitoring

Martin Tveten, Idris A. Eckley, Paul Fearnhead (2022) The Annals of Applied Statistics, 16(2), 721-743

Deep Learning for Anomaly Detection: A Review

Guansong Pang, Chunhua Shen, Longbing Cao, Anton Van Den Hengel (2021) ACM Computing Surveys (CSUR), 54(2), 1-38

High-Dimensional Time Series Segmentation via Factor-Adjusted Vector Autoregressive Modeling

Haeran Cho, Hyeyoung Maeng, Idris A. Eckley, Paul Fearnhead (2023) Journal of the American Statistical Association, 119(547), 2038–2050

On-Line Inference for Multiple Changepoint Problems

Paul Fearnhead, Zhen Liu (2007) Journal of the Royal Statistical Society Series B: Statistical Methodology, 69(4), 589–605

Tests for change-points with epidemic alternatives

Qiwei Yao (1993) Biometrika, 80(1), 179–191

A novel change-point approach for the detection of gas emission sources using remotely contained concentration data

Idris Eckley, Claudia Kirch, Silke Weber (2020) The Annals of Applied Statistics, 14(3), 1258-1284