Change Point Detection in Dynamic Graphs with Decoder-only Latent Space Model

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Kei, Yik Lun, Li, Jialiang, Li, Hangjian, Chen, Yanzhen, Padilla, Oscar Hernan Madrid
Format: Preprint
Publié: 2024
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866915246606647296
author Kei, Yik Lun
Li, Jialiang
Li, Hangjian
Chen, Yanzhen
Padilla, Oscar Hernan Madrid
author_facet Kei, Yik Lun
Li, Jialiang
Li, Hangjian
Chen, Yanzhen
Padilla, Oscar Hernan Madrid
contents This manuscript studies the unsupervised change point detection problem in time series of graphs using a decoder-only latent space model. The proposed framework consists of learnable prior distributions for low-dimensional graph representations and of a decoder that bridges the observed graphs and latent representations. The prior distributions of the latent spaces are learned from the observed data as empirical Bayes to assist change point detection. Specifically, the model parameters are estimated via maximum approximate likelihood, with a Group Fused Lasso regularization imposed on the prior parameters. The augmented Lagrangian is solved via Alternating Direction Method of Multipliers, and Langevin Dynamics are recruited for posterior inference. Simulation studies show good performance of the latent space model in supporting change point detection and real data experiments yield change points that align with significant events.
format Preprint
id arxiv_https___arxiv_org_abs_2404_04719
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Change Point Detection in Dynamic Graphs with Decoder-only Latent Space Model
Kei, Yik Lun
Li, Jialiang
Li, Hangjian
Chen, Yanzhen
Padilla, Oscar Hernan Madrid
Methodology
This manuscript studies the unsupervised change point detection problem in time series of graphs using a decoder-only latent space model. The proposed framework consists of learnable prior distributions for low-dimensional graph representations and of a decoder that bridges the observed graphs and latent representations. The prior distributions of the latent spaces are learned from the observed data as empirical Bayes to assist change point detection. Specifically, the model parameters are estimated via maximum approximate likelihood, with a Group Fused Lasso regularization imposed on the prior parameters. The augmented Lagrangian is solved via Alternating Direction Method of Multipliers, and Langevin Dynamics are recruited for posterior inference. Simulation studies show good performance of the latent space model in supporting change point detection and real data experiments yield change points that align with significant events.
title Change Point Detection in Dynamic Graphs with Decoder-only Latent Space Model
topic Methodology
url https://arxiv.org/abs/2404.04719