# Quickest Change-point Detection in Generalized Linear Models with Variational Estimation

arXiv: https://arxiv.org/abs/2610.16616 · HTML: https://mlsanity.com/paper/2610.16616/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 7 of its 21 references that are in the map.
- First posted on arXiv 12 Oct 2026.

## Builds on (most cited first, 7 of 7)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering](https://mlsanity.com/paper/2012.00955/) (2020)
- [Concept Drift Monitoring and Diagnostics of Supervised Learning Models via Score Vectors](https://mlsanity.com/paper/2012.06916/) (2020)
- [Window-Limited CUSUM for Sequential Change Detection](https://mlsanity.com/paper/2206.06777/) (2022)
- [Generalized generalized linear models: Convex estimation and online bounds](https://mlsanity.com/paper/2304.13793/) (2023)
- [Detecting Abrupt Changes in Sequential Pairwise Comparison Data](https://mlsanity.com/paper/2205.12431/) (2022)
- [Beyond Maximum Likelihood: Variational Inequality Estimation for Generalized Linear Models](https://mlsanity.com/paper/2511.03087/) (2025)

## Abstract (arXiv)

We consider quickest change-point detection in generalized linear models when the post-change parameter is unknown. Such problems arise, for example, when monitoring whether the relationship between a learned representation and an observed response changes over time. Since detection must occur soon after a change, the post-change parameter must be estimated from a limited number of observations, making finite-sample estimation accuracy particularly important. We propose a window-limited CUSUM procedure that estimates the post-change parameter using a variational-inequality (VI) formulation. For noncanonical links, VI estimation can offer favorable finite-sample accuracy over short windows despite the asymptotic efficiency advantage of maximum likelihood estimation. We characterize how estimation error reduces the effective post-change information rate and affects detection delay, and derive finite-window guarantees accounting for both statistical error and incomplete numerical optimization. The resulting procedure retains false-alarm control and achieves first-order asymptotic optimality under suitable conditions. Numerical experiments compare detection delay and update cost for fixed-budget estimation and benchmark VI-WLCUSUM against alternative detectors.