Kernel Change-point Detection with Auxiliary Deep Generative Models
This addresses a domain-specific challenge for researchers and practitioners in time series analysis, offering an incremental improvement by enhancing kernel two-sample tests with data-driven kernels.
The paper tackles the problem of kernel selection for change-point detection in time series by proposing KL-CPD, a kernel learning framework that uses an auxiliary generative model to optimize test power, resulting in significant outperformance over state-of-the-art methods in benchmark evaluations.
Detecting the emergence of abrupt property changes in time series is a challenging problem. Kernel two-sample test has been studied for this task which makes fewer assumptions on the distributions than traditional parametric approaches. However, selecting kernels is non-trivial in practice. Although kernel selection for two-sample test has been studied, the insufficient samples in change point detection problem hinder the success of those developed kernel selection algorithms. In this paper, we propose KL-CPD, a novel kernel learning framework for time series CPD that optimizes a lower bound of test power via an auxiliary generative model. With deep kernel parameterization, KL-CPD endows kernel two-sample test with the data-driven kernel to detect different types of change-points in real-world applications. The proposed approach significantly outperformed other state-of-the-art methods in our comparative evaluation of benchmark datasets and simulation studies.