Distill-Belief: Closed-Loop Inverse Source Localization and Characterization in Physical Fields

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shi, Yiwei, Song, Zixing, Yang, Mengyue, Liu, Cunjia, Liu, Weiru
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914515098009600
author Shi, Yiwei
Song, Zixing
Yang, Mengyue
Liu, Cunjia
Liu, Weiru
author_facet Shi, Yiwei
Song, Zixing
Yang, Mengyue
Liu, Cunjia
Liu, Weiru
contents {Closed-loop inverse source localization and characterization (ISLC) requires a mobile agent to select measurements that localize sources and infer latent field parameters under strict time constraints.} {The core challenge lies in the belief-space objective: valid uncertainty estimation requires expensive Bayesian inference, whereas using fast learned belief model leads to reward hacking, in which the policy exploits approximation errors rather than actually reducing uncertainty.} {We propose \textbf{Distill-Belief}, a teacher--student framework that decouples correctness from efficiency. A Bayes-correct particle-filter teacher maintains the posterior and supplies a dense information-gain signal, while a compact student distills the posterior into belief statistics for control and an uncertainty certificate for stopping. At deployment, only the student is used, yielding constant per-step cost.} {Experiments on seven field modalities and two stress tests show that Distill-Belief consistently reduces sensing cost and improves success, posterior contraction, and estimation accuracy over baselines, while mitigating reward hacking.}
format Preprint
id arxiv_https___arxiv_org_abs_2604_26095
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Distill-Belief: Closed-Loop Inverse Source Localization and Characterization in Physical Fields
Shi, Yiwei
Song, Zixing
Yang, Mengyue
Liu, Cunjia
Liu, Weiru
Artificial Intelligence
{Closed-loop inverse source localization and characterization (ISLC) requires a mobile agent to select measurements that localize sources and infer latent field parameters under strict time constraints.} {The core challenge lies in the belief-space objective: valid uncertainty estimation requires expensive Bayesian inference, whereas using fast learned belief model leads to reward hacking, in which the policy exploits approximation errors rather than actually reducing uncertainty.} {We propose \textbf{Distill-Belief}, a teacher--student framework that decouples correctness from efficiency. A Bayes-correct particle-filter teacher maintains the posterior and supplies a dense information-gain signal, while a compact student distills the posterior into belief statistics for control and an uncertainty certificate for stopping. At deployment, only the student is used, yielding constant per-step cost.} {Experiments on seven field modalities and two stress tests show that Distill-Belief consistently reduces sensing cost and improves success, posterior contraction, and estimation accuracy over baselines, while mitigating reward hacking.}
title Distill-Belief: Closed-Loop Inverse Source Localization and Characterization in Physical Fields
topic Artificial Intelligence
url https://arxiv.org/abs/2604.26095