Change detection with conformal martingales: new optimal constructions, and suboptimality of existing methods
Abstract
We study distribution-free sequential changepoint detection for independent observations with unknown and unrestricted pre- and post-change laws. We build on the conformal test martingales and associated e-detectors of Vovk(2021), which control the probability of false alarm (PFA) and the average run length (ARL) respectively. The majority of these works focus on validity, with statistical efficiency usually left for simulations. We develop a comprehensive theory of how conformal p-values behave under non-exchangeable data with a changepoint at an unknown time T. We use this to analyze the post-change growth and resulting detection delay of conformal martingale methods, and prove that the standard existing methods are suboptimal for PFA and ARL control, and can lead to delays that are Ω(T) and Ω(√(ARL)) respectively. We propose different conformal e-processes and e-detectors that are provably minimax optimal, with delays Θ(log T) and Θ(log ARL) respectively, and have much shorter delays in simulations.
Related
- Information-Geometric Blind Spots of Conformal Test Martingales
- Monitoring for Shifts at Unknown Times: Parameter-Free Conditional Conformal Martingales
- Reference-Null Calibrated Thresholds for E-Processes with Applications to Conformal Martingales
- A unified framework for existing and new constructions of Neumaier graphs
- Anytime-Valid Distribution Shift Detection via Predictive Rank Martingales
BibTeX
@misc{bhattacharyya2026change,
title = {{Change detection with conformal martingales: new optimal constructions, and suboptimality of existing methods}},
author = {Swapnaneel Bhattacharyya and Aaditya Ramdas},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.27179},
eprint = {2609.27179},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.27179},
}
RIS
TY - UNPB
TI - Change detection with conformal martingales: new optimal constructions, and suboptimality of existing methods
AU - Swapnaneel Bhattacharyya
AU - Aaditya Ramdas
PY - 2026
DA - 2026/09/23
PB - arXiv
DO - 10.48550/arxiv.2609.27179
UR - https://arxiv.org/abs/2609.27179
AB - We study distribution-free sequential changepoint detection for independent observations with unknown and unrestricted pre- and post-change laws. We build on the conformal test martingales and associated e-detectors of Vovk(2021), which control the probability of false alarm (PFA) and the average run length (ARL) respectively. The majority of these works focus on validity, with statistical efficiency usually left for simulations. We develop a comprehensive theory of how conformal p-values behave under non-exchangeable data with a changepoint at an unknown time T. We use this to analyze the post-change growth and resulting detection delay of conformal martingale methods, and prove that the standard existing methods are suboptimal for PFA and ARL control, and can lead to delays that are Ω(T) and Ω(√(ARL)) respectively. We propose different conformal e-processes and e-detectors that are provably minimax optimal, with delays Θ(log T) and Θ(log ARL) respectively, and have much shorter delays in simulations.
ER -