SSTLNetwork: a self-supervised spectral reconstruction network with hybrid attention for near-infrared spectral transfer learning

学习迁移 人工智能 遮罩(插图) 模式识别(心理学) 计算机科学 块(置换群论) 人工神经网络 过程(计算) 信号重构 化学 机器学习 监督学习 领域(数学分析) 三聚氰胺 灵活性(工程) 深度学习 迭代重建 传输(计算) 光谱形状分析 数据挖掘
作者
Lauren Gilman,H Wang,Nick Birse,Louise Manning
出处
期刊:Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy [Elsevier BV]
卷期号:361: 128082-128082
标识
DOI:10.1016/j.saa.2026.128082
摘要

Near-infrared (NIR) spectroscopy is widely used for non-destructive chemical analysis, yet models trained on one instrument or process condition often degrade when applied to data from different sources—a challenge known as domain shift. Transfer learning methods that rely on labelled target-domain data are not always practical in industrial or clinical settings. This work proposes SSTLNetwork, a self-supervised transfer learning approach for one-dimensional spectral data. The model employs a masked reconstruction pre-training objective with 75% patch masking and investigates a hybrid attention mechanism that pairs External Attention with multi-head self-attention. Pre-trained on a single melamine resin NIR dataset (3032 samples), SSTLNetwork is evaluated on held-out melamine recipe datasets, a pharmaceutical tablet dataset from a different spectrometer, and a virus classification task. After brief unsupervised fine-tuning (10 epochs), reconstruction on the tablet dataset achieves a mean absolute percentage error (MAPE) of 3.19 ± 1.04% and tolerance-based reconstruction accuracy (TA@0.01) of 86.55 ± 7.86%. Multi-seed ablation studies indicate that all architectural variants produce comparable reconstruction quality, suggesting that the masked reconstruction framework itself, rather than the specific attention configuration, is the primary driver of transfer learning performance. A preliminary downstream classification experiment on viral detection is also reported, though these results require further validation. These results, validated across five independent training runs, suggest that self-supervised masked reconstruction can yield transferable spectral representations, though further validation with additional datasets is warranted. • A self-supervised masked reconstruction framework adapted for one-dimensional NIR spectra. • A hybrid attention block combining External Attention and multi-head self-attention is investigated for capturing global and local spectral features. • Unsupervised fine-tuning on unlabelled target data enables cross-domain spectral reconstruction without reference values. • Multi-seed ablation studies reveal that the masked reconstruction framework drives transfer performance, with individual attention variants performing comparably. • A preliminary downstream classification experiment provides initial, suggestive evidence regarding representation transferability.
最长约 10秒,即可获得该文献文件

科研通智能强力驱动
Strongly Powered by AbleSci AI
科研通是完全免费的文献互助平台,具备全网最快的应助速度,最高的求助完成率。 对每一个文献求助,科研通都将尽心尽力,给求助人一个满意的交代。
实时播报
2秒前
3秒前
直率翠绿发布了新的文献求助10
3秒前
科研通AI6.4的应助被可爱采纳,获得10
3秒前
4秒前
孟语佳完成签到,获得积分10
4秒前
4秒前
5秒前
张伟发布了新的文献求助10
6秒前
6秒前
6秒前
6秒前
努力的宁完成签到,获得积分10
7秒前
布鲁鲁完成签到,获得积分10
7秒前
cyx发布了新的文献求助10
7秒前
ZIYUNI完成签到 ,获得积分10
9秒前
9秒前
cdercder的应助被皮卡丘采纳,获得10
10秒前
10秒前
Spice完成签到 ,获得积分10
11秒前
11秒前
FashionBoy的应助被leo采纳,获得10
11秒前
充电宝的应助被Prospect采纳,获得10
12秒前
lianmeiliu发布了新的文献求助10
12秒前
14秒前
yinyinyin发布了新的文献求助10
15秒前
16秒前
16秒前
17秒前
17秒前
17秒前
mengxi完成签到,获得积分10
18秒前
18秒前
向广宇发布了新的文献求助10
18秒前
19秒前
xxt完成签到,获得积分20
19秒前
无线那里是完成签到 ,获得积分10
19秒前
MANDY的应助被LBJ23采纳,获得10
19秒前
19秒前
Orange的应助被枯桃采纳,获得10
20秒前
高分求助中
(应助此贴封号)【重要!!请各用户(尤其是新用户)详细阅读】【科研通的精品贴汇总】 10000
拉伸高分子熔体中缠结行为的模拟和小角中子散射研究 1700
The World Atlas of Wine 9th edition 1000
Sex, Aging, and Death in a Medieval Medical Compendium: Trinity College Cambridge MS R.14.52, Its Texts, Language, and Scribe 500
高分子基功能界面水蒸发材料与应用研究 500
外场增强的水汽发电研究 500
基于结构调控的功能化水诱导发电材料与器件研究 500
热门求助领域 (近24小时)
化学 材料科学 医学 生物 纳米技术 计算机科学 工程类 有机化学 化学工程 生物化学 物理 复合材料 内科学 催化作用 光电子学 人工智能 细胞生物学 无机化学 物理化学 数学
热门帖子
关注 科研通微信公众号,转发送积分 8188221
求助须知:如何正确求助?哪些是违规求助? 9576025
关于积分的说明 21845049
捐赠科研通 7786165
什么是DOI,文献DOI怎么找? 3435226
关于科研通互助平台的介绍 2559559
邀请新用户注册赠送积分活动 2482157