How Many Posterior Samples? Calibrated Stopping for Adaptive Sensing
Abstract
In classification-oriented adaptive sensing, posterior samples characterize uncertainty at the current measurement state and can serve two roles: they may guide the next sensing direction, while their class labels provide votes for the candidate classes and determine whether sensing should continue. We focus on the stopping layer that turns these votes into a declaration, without modifying the posterior sampler or sensing directions. A natural plug-in rule declares when the observed vote share exceeds a threshold. We show that this threshold is not itself a confidence guarantee: when the underlying vote mass equals the threshold, the plug-in rule declares about half the time. As alternatives, we calibrate a fixed-sample rule and a finite-horizon sequential rule to a prescribed false-declaration probability, and study exact curtailment, which stops a fixed-pool rule once its final verdict is forced. We then derive how one-round declaration probabilities determine posterior-sample cost and classification accuracy along a sensing path. On MNIST with DDRM and a fixed PCA-guided probe sequence, curtailment saves up to 62% of posterior samples. Among the evaluated rules at matched operating points, sequential stopping reduces the cost the most. At a high accuracy, that same sequential rule can trade more posterior samples for fewer measurements.
Related
- SCOPE Before You Sample: Conformal Scores Guide Posterior Exploration for Fast and Calibrated Bayesian Predictive Inference
- Valid Stopping in Adaptive Generator-Verifier Loops
- Adaptive stopping and forecasting of screening, rescreening and randomization in multicentre clinical trials
- Tight Bounds for Equivalence Testing with Non-Adaptive Conditional Samples
- Recovery Guarantees for Posterior Sampling of One-Bit Compressed Sensing
BibTeX
@misc{corlay2026many,
title = {{How Many Posterior Samples? Calibrated Stopping for Adaptive Sensing}},
author = {Vincent Corlay and Andriy Enttsel},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2609.21813},
eprint = {2609.21813},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2609.21813},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - How Many Posterior Samples? Calibrated Stopping for Adaptive Sensing
AU - Vincent Corlay
AU - Andriy Enttsel
PY - 2026
DA - 2026/09/30
PB - arXiv
DO - 10.48550/arxiv.2609.21813
UR - https://arxiv.org/abs/2609.21813
AB - In classification-oriented adaptive sensing, posterior samples characterize uncertainty at the current measurement state and can serve two roles: they may guide the next sensing direction, while their class labels provide votes for the candidate classes and determine whether sensing should continue. We focus on the stopping layer that turns these votes into a declaration, without modifying the posterior sampler or sensing directions. A natural plug-in rule declares when the observed vote share exceeds a threshold. We show that this threshold is not itself a confidence guarantee: when the underlying vote mass equals the threshold, the plug-in rule declares about half the time. As alternatives, we calibrate a fixed-sample rule and a finite-horizon sequential rule to a prescribed false-declaration probability, and study exact curtailment, which stops a fixed-pool rule once its final verdict is forced. We then derive how one-round declaration probabilities determine posterior-sample cost and classification accuracy along a sensing path. On MNIST with DDRM and a fixed PCA-guided probe sequence, curtailment saves up to 62% of posterior samples. Among the evaluated rules at matched operating points, sequential stopping reduces the cost the most. At a high accuracy, that same sequential rule can trade more posterior samples for fewer measurements.
ER -