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

arXiv: https://arxiv.org/abs/2404.04719 · HTML: https://mlsanity.com/paper/2404.04719/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 6 Apr 2024.

## Built on by (most cited first, 1 of 1)

- [Accurate Evaluation of Quickest Changepoint Detectors via Non-parametric Survival Analysis](https://mlsanity.com/paper/2605.18798/) (2026)

## Abstract (arXiv)

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.