arXiv：cs.LG· Jawad Chowdhury, Ganesh Narasimha, Jan-Chi Yang, Hiroshi Funakubo, Yoshitaka Ehara, Yongtao Liu, Rama Vasudevan·· 7 小时前AI 评分35
基于高斯过程的质量控制主动学习：面向自主显微镜中鲁棒结构-性质学习
Quality-Controlled Active Learning via Gaussian Processes for Robust Structure-Property Learning in Autonomous Microscopy
AI 导读
研究者提出一种门控主动学习框架，将好奇心驱动采样与基于简谐振子模型拟合的物理信息质量控制过滤器结合，可在采集阶段自动排除低保真数据。在含空间局域噪声的 PbTiO3 薄膜 BEPS 数据上，该方法优于随机采样、标准主动学习和多任务学习策略，并在另一 PbTiO3 薄膜样品的实时自主显微镜实验中验证有效。
正文
Abstract:Autonomous experimental systems are increasingly used in materials research to accelerate scientific discovery, but their performance is often limited by low-quality, noisy data. This issue is especially problematic in data intensive structure-property learning tasks such as Image-to-Spectrum (Im2Spec) and Spectrum-to-Image (Spec2Im) translations, where standard active learning strategies can mistakenly prioritize poor quality measurements. We introduce a gated active learning framework that combines curiosity driven sampling with a physics-informed quality control filter based on Simple Harmonic Oscillator model fits, allowing the system to automatically exclude low fidelity data during acquisition. Evaluations on a pre-acquired dataset of band-excitation piezoresponse spectroscopy (BEPS) data from PbTiO3 thin films with spatially localized noise show that the proposed method outperforms random sampling, standard active learning, and multitask learning strategies. We further deployed the framework in real-time experiments on a separate PbTiO3 thin-film sample with heterogeneous domain structures, demonstrating its effectiveness in autonomous microscopy experiments. Overall, this work supports hybrid autonomy in self-driving labs, where physics-informed quality assessment and active decision making work together for more reliable discovery.
| Comments: | Published in npj Computational Materials (2026). Main text + Supplementary Information |
| Subjects: | Machine Learning (cs.LG) |
| Cite as: | arXiv:2603.29135 [cs.LG] |
| (or arXiv:2603.29135v3 [cs.LG] for this version) | |
| https://doi.org/10.48550/arXiv.2603.29135 arXiv-issued DOI via DataCite |
|
| Journal reference: | npj Computational Materials (2026) |
| Related DOI: | https://doi.org/10.1038/s41524-026-02248-x
DOI(s) linking to related resources |
Submission history
From: Md Hasan Jawad Chowdhury [view email]
[v1]
Tue, 31 Mar 2026 01:35:12 UTC (15,823 KB)
[v2]
Tue, 30 Jun 2026 02:51:26 UTC (15,823 KB)
[v3]
Mon, 5 Oct 2026 19:45:53 UTC (13,465 KB)
来源：arXiv：cs.LG · arxiv.org