Skip to yearly menu bar Skip to main content


Differentiable Change-point Detection With Temporal Point Processes

Paramita Koley · Harshavardhan Alimi · Shrey Singla · Sourangshu Bhattacharya · niloy ganguly · Abir De

Auditorium 1 Foyer 66


In this paper, we consider the problem of global change-point detection in event sequence data, where both the event distributions and change-points are assumed to be unknown. For this problem, we propose DCPD, a Log-likelihood Ratio based Global Change-point Detector, which detects change-points after observing the entire sequence. Based on the Transformer Hawkes Process (THP), a well-known neural TPP framework, we develop DCPD, a differentiable change-point detector, along with maintaining distinct intensity and mark predictor for each partition. Further, we propose DCPD-W, a sliding-window-based extension of DCPD to improve its scalability in terms of the number of events or change-points with minor sacrifice in performance. Experiments on synthetic datasets explore the effects of runtime, relative complexity, and other aspects of distributions on various properties of our change-point detector, namely robustness, detection accuracy, scalability, etc under controlled environments. Finally, we perform experiments on six real-world temporal event sequences collected from diverse domains like health, geographical regions, etc. We show that our methods either outperform or perform comparably with the baselines.

Live content is unavailable. Log in and register to view live content