Multinomial Sampling for Hierarchical Change-Point Detection
This work addresses a computational bottleneck in Bayesian change-point detection for applications like human behavior studies, though it appears incremental as it builds on existing methods with specific improvements.
The paper tackles the problem of uncertain point-estimates affecting change-point detection in high-dimensional time-series by proposing a multinomial sampling methodology, resulting in improved detection rates and reduced delay while maintaining stable complexity and analytical tractability.
Bayesian change-point detection, together with latent variable models, allows to perform segmentation over high-dimensional time-series. We assume that change-points lie on a lower-dimensional manifold where we aim to infer subsets of discrete latent variables. For this model, full inference is computationally unfeasible and pseudo-observations based on point-estimates are used instead. However, if estimation is not certain enough, change-point detection gets affected. To circumvent this problem, we propose a multinomial sampling methodology that improves the detection rate and reduces the delay while keeping complexity stable and inference analytically tractable. Our experiments show results that outperform the baseline method and we also provide an example oriented to a human behavior study.