Neural Networks最新文献_第4页

Hippocampus-centered structural covariance network reorganization in Alzheimer’s disease: An individualized graph-based biomarker validated by machine learning 阿尔茨海默病中以海马体为中心的结构协方差网络重组：通过机器学习验证的个体化基于图的生物标志物

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108542

Weiye Lu , Qian Gong , Yuna Chen , Shijun Qiu , Jie An

Alzheimer’s disease (AD) is characterized by progressive brain network disintegration, yet quantifying this process at an individual level remains challenging. This study explores the potential of an individualized differential structural covariance network (IDSCN) as a graph theory-based biomarker to capture disease-specific network reorganization. We found that throughout the AD spectrum, significant progressive atrophy occurred in multiple brain regions, especially the hippocampus. At the same time, the brain underwent a profound structural covariant reorganization, and this reorganization was significantly centered on the hippocampus. Graph theory analysis revealed a significant enhancement in nodal strength and nodal efficiency across widespread brain regions, with the hippocampus, amygdala, middle temporal gyrus, and entorhinal cortex emerging as core hubs of pathological impact. Importantly, betweenness centrality selectively increased only in the bilateral hippocampus, highlighting their critical role as bridges in the pathological propagation network. Machine learning validation confirmed that this individualized network biomarker performs excellently in distinguishing AD patients from cognitively normal individuals, demonstrates comparable efficacy to traditional morphological models in capturing early disease-related changes, and shows potential in differentiating between mild cognitive impairment converters and non-converters. Our study establishes the hippocampus-centered IDSCN as an effective, individualized graph theory-based biomarker, providing new insights into the network pathophysiology of AD and holding significant potential for early diagnosis and prognostic stratification.

阿尔茨海默病（AD）的特点是进行性脑网络解体，但在个体水平上量化这一过程仍然具有挑战性。本研究探讨了个体化差异结构协方差网络（IDSCN）作为基于图论的生物标志物捕捉疾病特异性网络重组的潜力。我们发现，在整个阿尔茨海默病谱系中，多个大脑区域，尤其是海马体，都出现了显著的进行性萎缩。与此同时，大脑发生了深刻的结构协变重组，这种重组主要集中在海马体。图论分析显示，在广泛的大脑区域中，节点强度和节点效率显著增强，海马、杏仁核、中颞回和内鼻皮层成为病理影响的核心枢纽。重要的是，中间性中心性仅在双侧海马中选择性增加，突出了它们在病理传播网络中作为桥梁的关键作用。机器学习验证证实，这种个性化的网络生物标志物在区分AD患者和认知正常个体方面表现出色，在捕捉早期疾病相关变化方面表现出与传统形态学模型相当的功效，并显示出区分轻度认知障碍转化者和非转化者的潜力。我们的研究确立了以海马体为中心的IDSCN是一种有效的、个性化的基于图论的生物标志物，为阿尔茨海默病的网络病理生理学提供了新的见解，并在早期诊断和预后分层方面具有重要潜力。

{"title":"Hippocampus-centered structural covariance network reorganization in Alzheimer’s disease: An individualized graph-based biomarker validated by machine learning","authors":"Weiye Lu , Qian Gong , Yuna Chen , Shijun Qiu , Jie An","doi":"10.1016/j.neunet.2026.108542","DOIUrl":"10.1016/j.neunet.2026.108542","url":null,"abstract":"<div><div>Alzheimer’s disease (AD) is characterized by progressive brain network disintegration, yet quantifying this process at an individual level remains challenging. This study explores the potential of an individualized differential structural covariance network (IDSCN) as a graph theory-based biomarker to capture disease-specific network reorganization. We found that throughout the AD spectrum, significant progressive atrophy occurred in multiple brain regions, especially the hippocampus. At the same time, the brain underwent a profound structural covariant reorganization, and this reorganization was significantly centered on the hippocampus. Graph theory analysis revealed a significant enhancement in nodal strength and nodal efficiency across widespread brain regions, with the hippocampus, amygdala, middle temporal gyrus, and entorhinal cortex emerging as core hubs of pathological impact. Importantly, betweenness centrality selectively increased only in the bilateral hippocampus, highlighting their critical role as bridges in the pathological propagation network. Machine learning validation confirmed that this individualized network biomarker performs excellently in distinguishing AD patients from cognitively normal individuals, demonstrates comparable efficacy to traditional morphological models in capturing early disease-related changes, and shows potential in differentiating between mild cognitive impairment converters and non-converters. Our study establishes the hippocampus-centered IDSCN as an effective, individualized graph theory-based biomarker, providing new insights into the network pathophysiology of AD and holding significant potential for early diagnosis and prognostic stratification.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108542"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145941312","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

AttCo: Attention-based co-Learning fusion of deep feature representation for medical image segmentation using multimodality 基于关注的深度特征表示融合多模态医学图像分割

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108572

Duy-Phuong Dao , Hyung-Jeong Yang , Soo-Hyung Kim , Sae-Ryung Kang

Accurate tissue segmentation is crucial for advancing healthcare, particularly in disease prediction and treatment planning. Precisely identifying abnormal tissue locations is a critical step for clinical analysis. While medical image segmentation increasingly utilizes multimodal and three-dimensional (3D) information to capture spatial relationships, current methods often struggle to effectively learn complementary information from multiple inputs, especially with complex 3D structures. In this study, we introduce AttCo, a novel multimodal semantic segmentation network built upon an attention-based co-learning fusion of deep feature representations. AttCo first employs multiple encoder branches to extract unimodal 3D representations from each imaging modality. These unimodal representations are then processed by a co-learning fusion module, which integrates both intra-modality (using SEAT) and inter-modality (using OSCAT) feature learning components. This dual approach ensures the capture of intricate interactions within each modality and across different modalities. Finally, these fused features pass through an up-sampling module to generate 3D segmented tumor maps. Our end-to-end deep network effectively addresses two key aspects: (i) extracting robust unimodal 3D representations and (ii) exploiting comprehensive inter- and intra-modality feature interactions. Experimental results demonstrate that AttCo significantly outperforms competing methods in terms of Dice score across various datasets. The source code can be found here https://github.com/duyphuongcri/AttCo.

准确的组织分割对于推进医疗保健至关重要，特别是在疾病预测和治疗计划方面。准确识别异常组织位置是临床分析的关键步骤。虽然医学图像分割越来越多地利用多模态和三维（3D）信息来捕捉空间关系，但目前的方法往往难以有效地从多个输入中学习互补信息，特别是复杂的3D结构。在这项研究中，我们引入了一种新的多模态语义分割网络AttCo，它建立在基于注意力的深度特征表征融合的共同学习基础上。AttCo首先使用多个编码器分支从每个成像模态中提取单模态3D表示。这些单模态表示然后由共同学习融合模块处理，该模块集成了模态内（使用SEAT）和模态间（使用OSCAT）特征学习组件。这种双重方法确保捕获每个模态内部和不同模态之间的复杂交互。最后，这些融合的特征通过上采样模块生成三维分割的肿瘤图。我们的端到端深度网络有效地解决了两个关键方面：(i)提取稳健的单模态3D表示和（ii）利用全面的模态间和模态内特征相互作用。实验结果表明，AttCo在不同数据集的Dice得分方面明显优于竞争方法。源代码可以在这里找到https://github.com/duyphuongcri/AttCo。

{"title":"AttCo: Attention-based co-Learning fusion of deep feature representation for medical image segmentation using multimodality","authors":"Duy-Phuong Dao , Hyung-Jeong Yang , Soo-Hyung Kim , Sae-Ryung Kang","doi":"10.1016/j.neunet.2026.108572","DOIUrl":"10.1016/j.neunet.2026.108572","url":null,"abstract":"<div><div>Accurate tissue segmentation is crucial for advancing healthcare, particularly in disease prediction and treatment planning. Precisely identifying abnormal tissue locations is a critical step for clinical analysis. While medical image segmentation increasingly utilizes multimodal and three-dimensional (3D) information to capture spatial relationships, current methods often struggle to effectively learn complementary information from multiple inputs, especially with complex 3D structures. In this study, we introduce AttCo, a novel multimodal semantic segmentation network built upon an attention-based co-learning fusion of deep feature representations. AttCo first employs multiple encoder branches to extract unimodal 3D representations from each imaging modality. These unimodal representations are then processed by a co-learning fusion module, which integrates both intra-modality (using SEAT) and inter-modality (using OSCAT) feature learning components. This dual approach ensures the capture of intricate interactions within each modality and across different modalities. Finally, these fused features pass through an up-sampling module to generate 3D segmented tumor maps. Our end-to-end deep network effectively addresses two key aspects: (i) extracting robust unimodal 3D representations and (ii) exploiting comprehensive inter- and intra-modality feature interactions. Experimental results demonstrate that AttCo significantly outperforms competing methods in terms of Dice score across various datasets. The source code can be found here <span><span>https://github.com/duyphuongcri/AttCo</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108572"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981716","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Attentional dual-stream interactive perception network for efficient infrared small aerial target detection 基于注意力双流交互感知网络的红外空中小目标高效检测

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108563

Lihao Zhou, Huawei Wang

Drones and other flying objects can be regarded as small targets from a long-distance perspective. Considering the occlusion and interference caused by the external environment, the infrared detection methods are adopted to help identify and manage small aerial targets. However, remote infrared imaging often leads to small target feature detail loss. And the general methods have low detection efficiency, difficult to deeply extract target features. To better address the above problems, we propose an attentional dual-stream interactive perception network (ADIPNet) in this paper. Based on dual-stream U-Net, ADIPNet mainly combines the multi-patch series-parallel attention module (MSPA), edge anchoring module with regret (EAR), context scene perception module (CSP) and dual-stream interaction fusion module (DSIF). MSPA manually constructs the weight of patch regions at multiple scales and then performs the nested self-attention so as to fully mine global target information. EAR unites two types of global features using local mapping and matrix product, which helps accurately capture small target edge. CSP exchanges context information multiple times and conducts mutual complementation of semantic scenarios to enhances the perception of small target features. Finally, DSIF conducts cross attention for high-level encoded features on double U-Nets, further improving the network’s understanding of complex scenario information. The proposed ADIPNet alleviates the insufficient feature extraction of infrared small targets. Compared with other state-of-the-art methods, mIoU respectively reaches 80.52% and 72.54% on two large infrared datasets. It achieves more accurate detection of small aerial targets with low operating cost, possessing potential application prospect in various infrared surveillance systems.

从远距离的角度来看，无人机和其他飞行物可以被视为小目标。考虑到外界环境的遮挡和干扰，采用红外探测方法对小型空中目标进行识别和管理。然而，远程红外成像往往会导致小目标特征细节的丢失。而一般方法检测效率低，难以深度提取目标特征。为了更好地解决上述问题，本文提出了一种注意力双流交互感知网络（ADIPNet）。ADIPNet基于双流U-Net，主要结合多补丁串并联注意模块（MSPA）、悔恨边缘锚定模块（EAR）、情境场景感知模块（CSP）和双流交互融合模块（DSIF）。MSPA在多个尺度上手动构建patch区域的权值，然后进行嵌套自关注，以充分挖掘全局目标信息。EAR利用局部映射和矩阵积将两种类型的全局特征结合起来，有助于准确捕获小目标边缘。CSP多次交换上下文信息，进行语义场景的相互补充，增强对小目标特征的感知。最后，DSIF对双U-Nets的高级编码特征进行交叉关注，进一步提高了网络对复杂场景信息的理解能力。所提出的ADIPNet缓解了红外小目标特征提取不足的问题。与其他最先进的方法相比，在两个大型红外数据集上mIoU分别达到80.52%和72.54%。该方法实现了对小型空中目标更精确的探测，运行成本低，在各种红外监视系统中具有潜在的应用前景。

{"title":"Attentional dual-stream interactive perception network for efficient infrared small aerial target detection","authors":"Lihao Zhou, Huawei Wang","doi":"10.1016/j.neunet.2026.108563","DOIUrl":"10.1016/j.neunet.2026.108563","url":null,"abstract":"<div><div>Drones and other flying objects can be regarded as small targets from a long-distance perspective. Considering the occlusion and interference caused by the external environment, the infrared detection methods are adopted to help identify and manage small aerial targets. However, remote infrared imaging often leads to small target feature detail loss. And the general methods have low detection efficiency, difficult to deeply extract target features. To better address the above problems, we propose an attentional dual-stream interactive perception network (ADIPNet) in this paper. Based on dual-stream U-Net, ADIPNet mainly combines the multi-patch series-parallel attention module (MSPA), edge anchoring module with regret (EAR), context scene perception module (CSP) and dual-stream interaction fusion module (DSIF). MSPA manually constructs the weight of patch regions at multiple scales and then performs the nested self-attention so as to fully mine global target information. EAR unites two types of global features using local mapping and matrix product, which helps accurately capture small target edge. CSP exchanges context information multiple times and conducts mutual complementation of semantic scenarios to enhances the perception of small target features. Finally, DSIF conducts cross attention for high-level encoded features on double U-Nets, further improving the network’s understanding of complex scenario information. The proposed ADIPNet alleviates the insufficient feature extraction of infrared small targets. Compared with other state-of-the-art methods, mIoU respectively reaches 80.52% and 72.54% on two large infrared datasets. It achieves more accurate detection of small aerial targets with low operating cost, possessing potential application prospect in various infrared surveillance systems.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108563"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981540","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Two-hidden-layer ReLU neural networks and finite elements 两隐层ReLU神经网络与有限元

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108559

Pengzhan Jin

We point out that (continuous or discontinuous) piecewise linear functions on a convex polytope mesh can be represented by two-hidden-layer ReLU neural networks in a weak sense. In addition, the numbers of neurons of the two hidden layers required to weakly represent are accurately given based on the numbers of polytopes and hyperplanes involved in this mesh. The results naturally hold for constant and linear finite element functions. Such weak representation establishes a bridge between two-hidden-layer ReLU neural networks and finite element functions, and leads to a perspective for analyzing approximation capability of ReLU neural networks in L^p norm via finite element functions. Moreover, we discuss the strict representation for tensor finite element functions via the recent tensor neural networks.

我们指出凸多面体网格上的（连续或不连续）分段线性函数可以用两隐层ReLU神经网络在弱意义上表示。此外，基于该网格中涉及的多面体和超平面的数量，精确地给出了弱表示所需的两个隐藏层的神经元数量。结果自然适用于常数和线性有限元函数。这种弱表示在两隐层ReLU神经网络和有限元函数之间架起了一座桥梁，为利用有限元函数分析ReLU神经网络在Lp范数中的逼近能力提供了一个视角。此外，我们还利用最近的张量神经网络讨论了张量有限元函数的严格表示。

引用次数: 0

Uncovering various neuronal responses in a fractional-order generalized HR system. 揭示分数阶广义HR系统中的各种神经元反应。

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108548

Krishnendu Bera, Chinmay Chakraborty, Eva Kaslik, Urszula Foryś, Sanjeev K Sharma, Argha Mondal

This study investigates neuronal electrical activities in a fractional-order generalized Hindmarsh-Rose (HR) system and explores an extended model incorporating an induced electric field. Stability and bifurcation analyses examine the impact of external electrical stimulation on neuronal dynamics. The results show how electric field parameters, including amplitude and frequency, modulate neuronal excitability and stability. The H-R model is a mathematical representation that captures diverse neuronal activities, and the introduction of fractional-order derivatives allows us to explore non-local dynamics in greater depth. We analyze the effects of fractional-order derivatives on the system's behavior, including the generation of action potential dynamics. We discuss some biophysical aspects of the different firing patterns that we encounter. In addition, the study employs both analytical and numerical methods to investigate the stability of bursting and spiking patterns, using linear stability analysis to examine the transitions between stable and unstable states. Simulations reveal significant memory effects even with a slight decrease in fractional order. This underscores the versatility of fractional-order models in bridging mathematical theory with biologically plausible phenomena. The findings of this study demonstrate the potential of fractional-order systems in capturing the intricacies of neuronal responses, highlighting the need for further exploration of these phenomena in excitable biophysical systems.

本文研究了分数阶广义Hindmarsh-Rose （HR）系统中的神经元电活动，并探索了一个包含感应电场的扩展模型。稳定性和分岔分析考察了外部电刺激对神经元动力学的影响。结果表明，电场参数（振幅和频率）对神经元的兴奋性和稳定性有调节作用。H-R模型是捕获不同神经元活动的数学表示，分数阶导数的引入使我们能够更深入地探索非局部动态。我们分析了分数阶导数对系统行为的影响，包括动作电位动力学的产生。我们讨论了我们遇到的不同放电模式的一些生物物理方面。此外，本研究采用解析和数值相结合的方法来研究爆裂和尖峰模式的稳定性，使用线性稳定性分析来研究稳定和不稳定状态之间的转换。模拟显示，即使分数阶略有下降，也会对记忆产生显著影响。这强调了分数阶模型在连接数学理论和生物学上似是而非的现象方面的多功能性。这项研究的发现证明了分数阶系统在捕捉神经元反应的复杂性方面的潜力，强调了在可兴奋生物物理系统中进一步探索这些现象的必要性。

{"title":"Uncovering various neuronal responses in a fractional-order generalized HR system.","authors":"Krishnendu Bera, Chinmay Chakraborty, Eva Kaslik, Urszula Foryś, Sanjeev K Sharma, Argha Mondal","doi":"10.1016/j.neunet.2026.108548","DOIUrl":"https://doi.org/10.1016/j.neunet.2026.108548","url":null,"abstract":"<p><p>This study investigates neuronal electrical activities in a fractional-order generalized Hindmarsh-Rose (HR) system and explores an extended model incorporating an induced electric field. Stability and bifurcation analyses examine the impact of external electrical stimulation on neuronal dynamics. The results show how electric field parameters, including amplitude and frequency, modulate neuronal excitability and stability. The H-R model is a mathematical representation that captures diverse neuronal activities, and the introduction of fractional-order derivatives allows us to explore non-local dynamics in greater depth. We analyze the effects of fractional-order derivatives on the system's behavior, including the generation of action potential dynamics. We discuss some biophysical aspects of the different firing patterns that we encounter. In addition, the study employs both analytical and numerical methods to investigate the stability of bursting and spiking patterns, using linear stability analysis to examine the transitions between stable and unstable states. Simulations reveal significant memory effects even with a slight decrease in fractional order. This underscores the versatility of fractional-order models in bridging mathematical theory with biologically plausible phenomena. The findings of this study demonstrate the potential of fractional-order systems in capturing the intricacies of neuronal responses, highlighting the need for further exploration of these phenomena in excitable biophysical systems.</p>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"108548"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145999305","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

HDFLStyler: Hierarchical domain-invariant feature learning for source-free domain generalization HDFLStyler：用于无源域泛化的分层域不变特征学习

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108546

Deqian Mao , Shanshan Gao , Faqiang Huang , Caiming Zhang , Yuanfeng Zhou

Source-Free Domain Generalization (SFDG) aims to generalize a model to unknown domains without using any specific source domain data. Currently, SFDG methods mainly use the vision-language large models to extract different style features from text prompts and use different style features to train linear classifiers, thereby eliminating the model’s dependence on source domain images. However, a challenging problem in the source-free domain generalization classification is how to generate as diverse styles as practicable solely from text prompts and learn domain-invariant features from different styles. In this paper, we propose a hierarchical domain-invariant feature learning method (HDFLStyler) for SFDG to improve the classification accuracy. HDFLStyler is mainly composed of diverse style generation and domain-invariant feature learning. Diverse style generation dynamically generates as many styles as practicable through random distribution adjustment and adaptive mixing strategies. Domain-invariant feature learning comprehensively learns domain-invariant features of diverse styles by combining global and local approaches. In addition, to better learn domain-invariant features, we also design a domain-invariant consistency loss to improve the classification performance of HDFLStyler. Extensive experiments demonstrate that HDFLStyler achieves excellent classification performance.

无源域泛化（SFDG）的目的是在不使用任何特定源域数据的情况下将模型泛化到未知域。目前，SFDG方法主要是利用视觉语言大模型从文本提示中提取不同的风格特征，并利用不同的风格特征训练线性分类器，从而消除了模型对源域图像的依赖。然而，在无源领域泛化分类中，一个具有挑战性的问题是如何仅从文本提示生成尽可能多样的样式，并从不同的样式中学习域不变特征。为了提高SFDG的分类精度，本文提出了一种分层域不变特征学习方法（HDFLStyler）。HDFLStyler主要由多样化风格生成和域不变特征学习两部分组成。多种风格生成通过随机分布调整和自适应混合策略动态生成尽可能多的风格。域不变特征学习是将全局和局部相结合的方法，对不同风格的域不变特征进行综合学习。此外，为了更好地学习域不变特征，我们还设计了一个域不变一致性损失来提高HDFLStyler的分类性能。大量的实验表明，HDFLStyler具有良好的分类性能。

{"title":"HDFLStyler: Hierarchical domain-invariant feature learning for source-free domain generalization","authors":"Deqian Mao , Shanshan Gao , Faqiang Huang , Caiming Zhang , Yuanfeng Zhou","doi":"10.1016/j.neunet.2026.108546","DOIUrl":"10.1016/j.neunet.2026.108546","url":null,"abstract":"<div><div>Source-Free Domain Generalization (SFDG) aims to generalize a model to unknown domains without using any specific source domain data. Currently, SFDG methods mainly use the vision-language large models to extract different style features from text prompts and use different style features to train linear classifiers, thereby eliminating the model’s dependence on source domain images. However, a challenging problem in the source-free domain generalization classification is how to generate as diverse styles as practicable solely from text prompts and learn domain-invariant features from different styles. In this paper, we propose a hierarchical domain-invariant feature learning method (HDFLStyler) for SFDG to improve the classification accuracy. HDFLStyler is mainly composed of diverse style generation and domain-invariant feature learning. Diverse style generation dynamically generates as many styles as practicable through random distribution adjustment and adaptive mixing strategies. Domain-invariant feature learning comprehensively learns domain-invariant features of diverse styles by combining global and local approaches. In addition, to better learn domain-invariant features, we also design a domain-invariant consistency loss to improve the classification performance of HDFLStyler. Extensive experiments demonstrate that HDFLStyler achieves excellent classification performance.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108546"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981633","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel cross-domain fault diagnosis method for multi-condition industrial processes based on meta-domain adaptation with progressive meta-learning 基于元域自适应和渐进式元学习的多工况工业过程跨域故障诊断方法。

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108561

Xin Qin , Xuan Guo , Jie Dong , Kaixiang Peng

Complex industrial processes are characterized by high dynamics, diverse operating conditions, and strong inter-system coupling, often leading to reduced production efficiency and product quality fluctuations. Employing advanced fault diagnosis technologies has become an effective approach to support high-quality and efficient execution of industrial processes. However, the increasing prevalence of customized manufacturing has introduced substantial variability in working conditions, under which traditional fault diagnosis methods struggle to perform effectively. Each working condition can be abstracted as a domain. Therefore, employing domain adaptation techniques to achieve multi-condition fault diagnosis is one of the key approaches to addressing the above challenge. Based on the above observation, a novel neural network-based cross-domain fault diagnosis method for multi-condition industrial processes via meta-domain adaptation with progressive meta-learning is proposed. First, an adversarial dual-scale neural network is designed to address the challenge of feature alignment across multiple source domains, comprising a one-dimensional convolutional neural network feature extractor and a multi-layer perceptrons domain discriminator. A progressive adversarial strength adjustment strategy is proposed to better extract domain-invariant yet discriminative shared features, thereby enhancing domain generalization. Second, to tackle practical issues such as imbalanced condition distributions, limited sample availability, and intra-source heterogeneity, a meta-learning mechanism is employed to reduce internal distributional discrepancies within source domains. Additionally, multi-kernel maximum mean discrepancy is employed to explicitly align source and target features, facilitating robust generalization under substantial domain shifts. Finally, the constructed cross-domain feature extractor and fault classifier are used to achieve fault diagnosis in industrial processes. The proposed method is evaluated on the benchmark Tennessee Eastman process and a real hot strip mill process, demonstrating its effectiveness and superiority.

复杂的工业过程具有高动态性、多样化的操作条件和强系统间耦合的特点，往往导致生产效率降低和产品质量波动。采用先进的故障诊断技术已成为支持高质量和高效执行工业过程的有效途径。然而，定制制造的日益普及带来了工作条件的实质性变化，在这种情况下，传统的故障诊断方法难以有效地发挥作用。每个工况都可以抽象为一个域。因此，利用领域自适应技术实现多条件故障诊断是解决上述问题的关键途径之一。在此基础上，提出了一种基于神经网络的多工况工业过程跨域故障诊断方法，该方法基于元域自适应和渐进式元学习。首先，设计了一种对抗双尺度神经网络来解决跨多源域特征对齐的挑战，包括一维卷积神经网络特征提取器和多层感知器域鉴别器。提出了一种渐进式对抗强度调整策略，以更好地提取域不变但有区别的共享特征，从而提高域泛化能力。其次，为了解决条件分布不平衡、样本可用性有限和源内异质性等实际问题，采用元学习机制减少源域内部分布差异。此外，采用多核最大平均差异来显式对齐源和目标特征，便于在大量域偏移下进行鲁棒泛化。最后，利用构建的跨域特征提取器和故障分类器实现工业过程的故障诊断。通过田纳西州伊士曼工艺和热轧带钢实际工艺对该方法进行了评价，验证了该方法的有效性和优越性。

{"title":"A novel cross-domain fault diagnosis method for multi-condition industrial processes based on meta-domain adaptation with progressive meta-learning","authors":"Xin Qin , Xuan Guo , Jie Dong , Kaixiang Peng","doi":"10.1016/j.neunet.2026.108561","DOIUrl":"10.1016/j.neunet.2026.108561","url":null,"abstract":"<div><div>Complex industrial processes are characterized by high dynamics, diverse operating conditions, and strong inter-system coupling, often leading to reduced production efficiency and product quality fluctuations. Employing advanced fault diagnosis technologies has become an effective approach to support high-quality and efficient execution of industrial processes. However, the increasing prevalence of customized manufacturing has introduced substantial variability in working conditions, under which traditional fault diagnosis methods struggle to perform effectively. Each working condition can be abstracted as a domain. Therefore, employing domain adaptation techniques to achieve multi-condition fault diagnosis is one of the key approaches to addressing the above challenge. Based on the above observation, a novel neural network-based cross-domain fault diagnosis method for multi-condition industrial processes via meta-domain adaptation with progressive meta-learning is proposed. First, an adversarial dual-scale neural network is designed to address the challenge of feature alignment across multiple source domains, comprising a one-dimensional convolutional neural network feature extractor and a multi-layer perceptrons domain discriminator. A progressive adversarial strength adjustment strategy is proposed to better extract domain-invariant yet discriminative shared features, thereby enhancing domain generalization. Second, to tackle practical issues such as imbalanced condition distributions, limited sample availability, and intra-source heterogeneity, a meta-learning mechanism is employed to reduce internal distributional discrepancies within source domains. Additionally, multi-kernel maximum mean discrepancy is employed to explicitly align source and target features, facilitating robust generalization under substantial domain shifts. Finally, the constructed cross-domain feature extractor and fault classifier are used to achieve fault diagnosis in industrial processes. The proposed method is evaluated on the benchmark Tennessee Eastman process and a real hot strip mill process, demonstrating its effectiveness and superiority.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108561"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145967132","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks 一种基于负峰度损失的反向传播算法，用于训练浅层、卷积和深层神经网络

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108570

Engin Cemal Mengüç , Alper Emlek , Danilo P. Mandic

The conventional backpropagation (BP) algorithm remains the most widely used approach for training neural networks (NNs), including shallow NN (SNN), convolutional NN (CNN), deep NN (DNN), and deep CNN (DCNN), due to its easy implementation and well-described mathematical structure. However, the BP algorithm has several drawbacks, such as slow convergence and high steady-state error arising from its mean square error (MSE) loss function. For this reason, in this paper, we propose a novel BP algorithm to improve the training and testing efficiencies of SNN, CNN, DNN, and DCNN architectures. This is achieved by minimizing a loss function defined as the negated kurtosis of the error of the output layer. Moreover, the proposed kurtosis-based BP algorithm, which originally employs stochastic gradient descent (SGD), is extended to incorporate more advanced optimizers, such as root mean square propagation (RMSProp) and adaptive moment estimation (Adam). The experimental results on the regression and classification problems indicate that, in the training and testing procedures of all the NN architectures, the proposed kurtosis-based BP algorithm not only increases the convergence rate but also decreases the steady-state error when compared to the well-known competitive methods.

传统的反向传播（BP）算法仍然是训练神经网络（NN）最广泛使用的方法，包括浅层神经网络（SNN）、卷积神经网络（CNN）、深度神经网络（DNN）和深度神经网络（DCNN），因为它易于实现和描述良好的数学结构。然而，BP算法存在收敛速度慢、均方误差（MSE）损失函数导致稳态误差大等缺点。因此，在本文中，我们提出了一种新的BP算法来提高SNN、CNN、DNN和DCNN架构的训练和测试效率。这是通过最小化定义为输出层误差的负峰度的损失函数来实现的。此外，本文提出的基于峰度的BP算法，最初采用随机梯度下降（SGD），扩展到更先进的优化器，如均方根传播（RMSProp）和自适应矩估计（Adam）。在回归和分类问题上的实验结果表明，在所有神经网络架构的训练和测试过程中，与已知的竞争方法相比，本文提出的基于峭度的BP算法不仅提高了收敛速度，而且减小了稳态误差。

{"title":"A novel backpropagation algorithm based on negated kurtosis loss for training shallow, convolutional, and deep neural networks","authors":"Engin Cemal Mengüç , Alper Emlek , Danilo P. Mandic","doi":"10.1016/j.neunet.2026.108570","DOIUrl":"10.1016/j.neunet.2026.108570","url":null,"abstract":"<div><div>The conventional backpropagation (BP) algorithm remains the most widely used approach for training neural networks (NNs), including shallow NN (SNN), convolutional NN (CNN), deep NN (DNN), and deep CNN (DCNN), due to its easy implementation and well-described mathematical structure. However, the BP algorithm has several drawbacks, such as slow convergence and high steady-state error arising from its mean square error (MSE) loss function. For this reason, in this paper, we propose a novel BP algorithm to improve the training and testing efficiencies of SNN, CNN, DNN, and DCNN architectures. This is achieved by minimizing a loss function defined as the negated kurtosis of the error of the output layer. Moreover, the proposed kurtosis-based BP algorithm, which originally employs stochastic gradient descent (SGD), is extended to incorporate more advanced optimizers, such as root mean square propagation (RMSProp) and adaptive moment estimation (Adam). The experimental results on the regression and classification problems indicate that, in the training and testing procedures of all the NN architectures, the proposed kurtosis-based BP algorithm not only increases the convergence rate but also decreases the steady-state error when compared to the well-known competitive methods.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108570"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Temporal local attention with adaptive decoding: Enhancing spiking neural networks for temporal computing applications 具有自适应解码的时间局部注意：用于时间计算应用的增强尖峰神经网络

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-08 DOI: 10.1016/j.neunet.2026.108558

Hanxiao Fan , Hanle Zheng , Zikai Wang , Jiayi Mao , Huifeng Yin , Hao Guo , Lei Deng

The brain-inspired spiking neural networks (SNNs) are considered to have great potential in complex learning due to their rich neural dynamics and high energy efficiency. Their unique mechanisms are naturally suited for performing temporal computing tasks. However, whether SNNs can effectively capture sufficient temporal features solely based on the neural dynamics remains to be explored, as they usually encounter difficulties in long-sequence gradient propagation during training. In this work, we introduce temporal local attention (TLA), which helps SNNs effectively reduce the sequence length, thereby enhancing the model performance and accelerating the training process. Furthermore, we optimize the output process of SNNs by incorporating adaptive decoding (AD) based on finding the correlation between the decoding strategy and model performance. Finally, we combine these two mechanisms (TLA-AD) into complete SNN modeling with training and inference. We use LSTM and a recent SNN model with dendritic heterogeneity (DH-SNN) as baselines on Electroencephalogram (EEG) and natural language processing (NLP) datasets. The experimental results demonstrate that the proposed TLA-AD method significantly enhances the performance of SNNs and accelerates the training without a significant increase of the number of parameters. It achieves state-of-the-art accuracy scores of 93.52% and 93.41% on the DEAP dataset (valence and arousal), as well as competitive accuracy scores of 87.41% and 86.31% on SEED and IMDB datasets, respectively, compared to other SNN methods. This work provides effective optimization approaches for enhancing SNNs with advanced performance in temporal computing tasks.

大脑激发的脉冲神经网络（SNNs）由于其丰富的神经动力学和高能量效率而被认为在复杂学习中具有巨大的潜力。它们独特的机制自然适合于执行时间计算任务。然而，由于snn在训练过程中通常会遇到长序列梯度传播的困难，因此仅基于神经动力学的snn能否有效捕获足够的时间特征还有待探索。在这项工作中，我们引入了时间局部注意（temporal local attention， TLA），它可以帮助snn有效地减少序列长度，从而提高模型性能并加速训练过程。此外，我们在发现解码策略与模型性能之间的相关性的基础上，通过引入自适应解码（AD）来优化snn的输出过程。最后，我们将这两种机制（TLA-AD）结合起来，形成完整的SNN模型，并进行训练和推理。我们使用LSTM和最近的具有树突异质性的SNN模型（DH-SNN）作为脑电图（EEG）和自然语言处理（NLP）数据集的基线。实验结果表明，提出的TLA-AD方法在不显著增加参数数量的情况下，显著提高了snn的性能，加快了训练速度。与其他SNN方法相比，该方法在DEAP数据集（效价和唤醒）上的准确率分别为93.52%和93.41%，在SEED和IMDB数据集上的竞争准确率分别为87.41%和86.31%。这项工作为增强snn在时间计算任务中的高级性能提供了有效的优化方法。

{"title":"Temporal local attention with adaptive decoding: Enhancing spiking neural networks for temporal computing applications","authors":"Hanxiao Fan , Hanle Zheng , Zikai Wang , Jiayi Mao , Huifeng Yin , Hao Guo , Lei Deng","doi":"10.1016/j.neunet.2026.108558","DOIUrl":"10.1016/j.neunet.2026.108558","url":null,"abstract":"<div><div>The brain-inspired spiking neural networks (SNNs) are considered to have great potential in complex learning due to their rich neural dynamics and high energy efficiency. Their unique mechanisms are naturally suited for performing temporal computing tasks. However, whether SNNs can effectively capture sufficient temporal features solely based on the neural dynamics remains to be explored, as they usually encounter difficulties in long-sequence gradient propagation during training. In this work, we introduce temporal local attention (TLA), which helps SNNs effectively reduce the sequence length, thereby enhancing the model performance and accelerating the training process. Furthermore, we optimize the output process of SNNs by incorporating adaptive decoding (AD) based on finding the correlation between the decoding strategy and model performance. Finally, we combine these two mechanisms (TLA-AD) into complete SNN modeling with training and inference. We use LSTM and a recent SNN model with dendritic heterogeneity (DH-SNN) as baselines on Electroencephalogram (EEG) and natural language processing (NLP) datasets. The experimental results demonstrate that the proposed TLA-AD method significantly enhances the performance of SNNs and accelerates the training without a significant increase of the number of parameters. It achieves state-of-the-art accuracy scores of 93.52% and 93.41% on the DEAP dataset (valence and arousal), as well as competitive accuracy scores of 87.41% and 86.31% on SEED and IMDB datasets, respectively, compared to other SNN methods. This work provides effective optimization approaches for enhancing SNNs with advanced performance in temporal computing tasks.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108558"},"PeriodicalIF":6.3,"publicationDate":"2026-01-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145981727","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Kernelized linear principal component discriminant analysis 核化线性主成分判别分析。

IF 6.3 1区计算机科学 Q1 COMPUTER SCIENCE, ARTIFICIAL INTELLIGENCE

Neural Networks

Pub Date : 2026-01-07 DOI: 10.1016/j.neunet.2026.108539

Lingxiao Qu , Yan Pei

In this paper, we propose Kernelized Linear Principal Component Discriminant Analysis (KLPCDA), a structured and unified framework for discriminant analysis that overcomes the fragmentation in existing multi-stage approaches such as PCA+LDA or KPCA+GDA. Instead of treating feature extraction and class discrimination as disjoint steps, KLPCDA formulates a joint optimization model in the Reproducing Kernel Hilbert Space (RKHS), integrating overall variance preservation, between-class separation, and within-class compactness into a fused objective. The formulation supports seven KLPCDA variants, offering flexible control over effects of each criterion through tunable fusion coefficients. We present a systematic parameter optimization strategy, including kernel parameter selection, subspace dimensionality tuning, and fusion balancing, along with an alternative kernel parameter optimization method. Extensive experiments across image, tabular, and signal datasets across small-sample-size (SSS) to larger-scale settings validate the adaptability of KLPCDA. The results demonstrate KLPCDA consistently outperforms benchmark methods and convolutional neural networks in SSS settings on both recognition accuracy and efficiency, while maintaining competitive advantages in computational complexity and storage requirements for large-scale scenarios. Finally, we provide insights into extended study subjects and future work related to our proposal.

本文提出了一种结构化、统一的判别分析框架——核化线性主成分判别分析（KLPCDA），克服了PCA+LDA或KPCA+GDA等多阶段判别分析方法的碎片化问题。KLPCDA没有将特征提取和类判别作为不相关的步骤，而是在再现核希尔伯特空间（RKHS）中建立了一个联合优化模型，将总体方差保持、类间分离和类内紧密性整合到一个融合目标中。该配方支持七个KLPCDA变体，通过可调的融合系数提供灵活的控制每个标准的影响。我们提出了一种系统的参数优化策略，包括核参数选择、子空间维数调整和融合平衡，以及一种替代的核参数优化方法。从小样本（SSS）到大规模设置的图像、表格和信号数据集的广泛实验验证了KLPCDA的适应性。结果表明，在SSS设置下，KLPCDA在识别精度和效率方面始终优于基准方法和卷积神经网络，同时在大规模场景的计算复杂度和存储要求方面保持竞争优势。最后，我们提供了与我们的提案相关的扩展研究主题和未来工作的见解。

{"title":"Kernelized linear principal component discriminant analysis","authors":"Lingxiao Qu , Yan Pei","doi":"10.1016/j.neunet.2026.108539","DOIUrl":"10.1016/j.neunet.2026.108539","url":null,"abstract":"<div><div>In this paper, we propose Kernelized Linear Principal Component Discriminant Analysis (KLPCDA), a structured and unified framework for discriminant analysis that overcomes the fragmentation in existing multi-stage approaches such as PCA+LDA or KPCA+GDA. Instead of treating feature extraction and class discrimination as disjoint steps, KLPCDA formulates a joint optimization model in the Reproducing Kernel Hilbert Space (RKHS), integrating overall variance preservation, between-class separation, and within-class compactness into a fused objective. The formulation supports seven KLPCDA variants, offering flexible control over effects of each criterion through tunable fusion coefficients. We present a systematic parameter optimization strategy, including kernel parameter selection, subspace dimensionality tuning, and fusion balancing, along with an alternative kernel parameter optimization method. Extensive experiments across image, tabular, and signal datasets across small-sample-size (SSS) to larger-scale settings validate the adaptability of KLPCDA. The results demonstrate KLPCDA consistently outperforms benchmark methods and convolutional neural networks in SSS settings on both recognition accuracy and efficiency, while maintaining competitive advantages in computational complexity and storage requirements for large-scale scenarios. Finally, we provide insights into extended study subjects and future work related to our proposal.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"198 ","pages":"Article 108539"},"PeriodicalIF":6.3,"publicationDate":"2026-01-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145967199","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0