Neural Networks最新文献

{"title":"Loyalty-SMOTE: Data synthesis algorithm for effective imbalanced data classification","authors":"Shengquan Hu ,&nbsp;Junfei Li ,&nbsp;Zefeng Li ,&nbsp;Zihao Zhang ,&nbsp;Yan Feng ,&nbsp;K. L Eddie Law","doi":"10.1016/j.neunet.2026.108677","DOIUrl":"10.1016/j.neunet.2026.108677","url":null,"abstract":"<div><div>Imbalanced datasets are always problematic in training machine learning models, so that classifiers often struggle to achieve satisfactory performance. Numerous approaches have been developed to tackle imbalanced data problems. Among them, some data-level methods perform linear interpolations between neighboring minority class samples to generate new data points, while others focus on oversampling boundary samples which are specific to certain classes. However, many methods fail to consider scenarios involving noise susceptibility. In this paper, we propose a novel data-level method called the <em>Loyalty-SMOTE</em> algorithm. We introduce the concept of <em>Loyalty</em> to identify noise and boundaries within datasets. After identifying potential noisy datapoints, SMOTE (Synthetic Minority Oversampling Technique) algorithm is applied to oversample the minority class boundary data. Subsequently, a denoising process based on Loyalty is conducted to obtain a balanced dataset. To extend our design, the concept of <em>Attraction</em> is introduced to generalize the denoising technique for multiclass problems. In our study, the SVM (Support Vector Machine) classifier is used as our base learner,and extensive experiments are performed to evaluate and compare different algorithms. Our results demonstrate that Loyalty-SMOTE achieved superior performance across multiple metrics on both binary and multiclass UCI datasets. For 30 binary datasets, it achieved the highest scores in 26 datasets (87%) for F1-score, 29 datasets (97%) for AUROC, 26 datasets (87%) for recall, and 27 datasets (90%) for G-mean. For 5 multiclass datasets, our design achieved scores of 0.8317, 0.6153, 0.8537, and 0.6717, respectively.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108677"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146158915","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}

{"title":"A dimensional structure based knowledge distillation method for cross-modal learning","authors":"Lingyu Si ,&nbsp;Hongwei Dong ,&nbsp;Fan Yang ,&nbsp;Shouyou Huang ,&nbsp;Junzhi Yu ,&nbsp;Fuchun Sun","doi":"10.1016/j.neunet.2026.108672","DOIUrl":"10.1016/j.neunet.2026.108672","url":null,"abstract":"<div><div>Transferring informative dark knowledge from other modalities has become a common approach to solving learning tasks that are challenging to accomplish independently due to limitations in data quality. However, research on why the transferred knowledge works has not been extensively explored. To address this issue, in this paper, we discover the correlation between feature discriminability and its dimensional structure (DS) by observing the features extracted from modalities with high and low data quality within the same learning task. On this basis, we express DS using the spatial distribution of intermediate features and the channel-wise correlation of output features. We empirically find that the DS of high-quality features is better than that of low-quality ones. This inspires us to propose a novel DS-based knowledge distillation method for better supervised cross-modal learning (CML) performance. Instead of merely mimicking the logits or features from the high-quality modality, the proposed method leverages its structural knowledge to guide the low-quality modality. Specifically, it enforces uniform distribution of intermediate features and channel-wise independence of deep features in the low-quality modality, thereby enhancing semantic learning and improving performance. This is especially useful when the performance gap between dual modalities is relatively large. Furthermore, this paper introduces a new CML dataset for the task of marine target recognition, named IIS-ISCAS, to promote community development. The dataset includes more than 10,000 paired samples from 8 distinct marine targets with optics and radar modalities and is continuously being updated. Experimental results on six visual benchmark transformed datasets and six CML datasets validate the effectiveness of the proposed method.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108672"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146158719","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}

{"title":"Structure-missing graph-level clustering network","authors":"Tianyu Hu ,&nbsp;Renda Han ,&nbsp;Liu Mao ,&nbsp;Jing Chen ,&nbsp;Xia Xie","doi":"10.1016/j.neunet.2026.108682","DOIUrl":"10.1016/j.neunet.2026.108682","url":null,"abstract":"<div><div>Graph-level clustering aims to group graphs into distinct clusters based on shared structural characteristics or semantic similarities. However, existing graph-level clustering methods generally assume that the input graph structure is complete and overlook the problem of missing relationships that commonly exist in real-world scenarios. These unmodeled missing relationships will lead to the accumulation of structural information distortion during the graph representation learning process, significantly reducing the clustering performance. To this end, we propose a novel method, Structure-Missing Graph-Level Clustering Network (SMGCN), which includes a structure augmentation module LR-SEA, an Anchor Positioning Mechanism, and Joint Contrastive Optimization. Specifically, we first output augmented graphs based on low-rank matrix completion, perform cluster matching using the Hungarian algorithm to obtain anchors, and then force same clustering graphs to converge to the corresponding anchors in the embedding space. According to our research, this is the first time that the graph-level clustering task with missing relations is proposed, and the superiority of our method is demonstrated through experiments on five benchmark datasets, compared with the state-of-the-art methods. Our source codes are available at <span><span>https://github.com/MrHuSN/SMGCN</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108682"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146144622","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}

{"title":"MCDNet: Morphological-conditional dual-view fusion for 3D tubular structure segmentation","authors":"Zhiyan Wang ,&nbsp;Changjian Wang ,&nbsp;Kele Xu ,&nbsp;Zhongshun Tang ,&nbsp;Yan Zhuang ,&nbsp;Jiani Zou ,&nbsp;Fangyi Liu","doi":"10.1016/j.neunet.2026.108614","DOIUrl":"10.1016/j.neunet.2026.108614","url":null,"abstract":"<div><div>Accurate segmentation of 3D tubular structures in medical images is critical for clinical diagnosis and interventional planning. Although deep learning methods have advanced significantly, most existing approaches exhibit limited generalizability due to their reliance on structure-specific morphological priors. Consequently, these models are often constrained to particular anatomical regions–such as the colorectal tract, vasculature, or arteries–leading to suboptimal performance when applied across varied organ systems. Moreover, the joint modeling of global and local morphological characteristics remains underexplored in the context of tubular structure segmentation. To address these limitations, we propose MCDNet, a Morphological-Conditional Dual-view Network that integrates both contextual and morphological information through a target-adaptive Morphological-Conditional Convolution (MCConv). The network is composed of three sequential modules: (1) a morphological feature extraction stage leveraging MCConv to enhance structural sensitivity across diverse tubular geometries; (2) a contextual feature learning stage employing a cross-fusion mechanism that synergistically combines convolutional and attention-based representations; and (3) a residual self-attention fusion module that reinforces feature integration from decoupled morphological and contextual branches. We evaluate MCDNet on four benchmark datasets encompassing diverse tubular segmentation tasks. Experimental results demonstrate that MCDNet achieves superior performance over state-of-the-art methods, with an average Dice coefficient improvement of 6.93% and a 10.61% reduction in Hausdorff distance relative to a strong baseline. The source code is publicly available at <span><span>https://github.com/wzydcg/MCDNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108614"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081675","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}

{"title":"A knowledge-driven self-supervised learning method for enhancing EEG-based emotion recognition","authors":"Hanqi Wang ,&nbsp;Jingyu Zhang ,&nbsp;Peng Ye ,&nbsp;Kun Yang ,&nbsp;Jichuan Xiong ,&nbsp;Xuefeng Liu ,&nbsp;Tao Chen ,&nbsp;Liang Song","doi":"10.1016/j.neunet.2026.108676","DOIUrl":"10.1016/j.neunet.2026.108676","url":null,"abstract":"<div><div>Emotion recognition brain-computer interface (BCI) using electroencephalography (EEG) is crucial for human-computer interaction, medicine, and neuroscience. However, the scarcity of labeled EEG data limits progress in this field. To address this, self-supervised learning has gained attention as a promising approach. Despite its potential, self-supervised methods face two key challenges: (1) ensuring emotion-related information is effectively preserved, as its loss can degrade emotion recognition performance, and (2) overcoming inter-subject variability in EEG signals, which hinders generalization across subjects. To tackle these issues, we propose a novel knowledge-driven self-supervised learning framework for EEG emotion recognition. Our method incorporates domain knowledge to approximate the extraction of statistical feature differential entropy (DE), aiming to preserve emotion-related and generalizable information. The framework consists of two cascaded components as hard and soft alignments: a multi-branch convolutional differential entropy learning (MCDEL) module that simulates the DE extraction process, and a contrastive entropy alignment (CEA) module that exposes complex emotional semantics in high-dimensional space. Experiment results show that our method exhibits superior performance over existing self-supervised methods. The subject-independent mean accuracy and standard deviation of our method reached 84.48% ± 5.79 on SEED and 67.64% ± 6.35 and 68.63% ± 7.77 on the Arousal and Valence dimensions of DREAMER, respectively. We conduct an ablation study to demonstrate the contribution of each proposed component. Moreover, the t-SNE visualization intuitively presents the effect of our method on reducing inter-subject variability and discriminating emotional states.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108676"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146158637","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}

{"title":"Adversarially robust neural network decision boundaries via tropical geometry","authors":"Kurt Pasque ,&nbsp;Christopher Teska ,&nbsp;Ruriko Yoshida ,&nbsp;Keiji Miura ,&nbsp;Jefferson Huang","doi":"10.1016/j.neunet.2026.108624","DOIUrl":"10.1016/j.neunet.2026.108624","url":null,"abstract":"<div><div>We introduce a simple, easy to implement, and computationally efficient tropical convolutional neural network architecture that is robust against adversarial attacks. We exploit the tropical nature of piece-wise linear neural networks by embedding the data in the <em>tropical projective torus</em>. This can be accomplished with a single additional hidden layer called a <em>tropical embedding layer</em>, and can in principle be added to any neural network architecture. We study the geometry of the resulting decision boundary, and find that like adversarial training and various regularization techniques that have been proposed, adding the tropical embedding layer tends to increase the number of linear regions associated with the decision boundaries. Our numerical experiments show that our approach achieves state-of-the-art levels of adversarial robustness, while requiring much less computational time than adversarial training.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108624"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081762","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}

{"title":"Cross-view contrastive representation learning on meta-path induced graphs with node features for bundle recommendation","authors":"Peng Zhang ,&nbsp;Zhendong Niu ,&nbsp;Ru Ma ,&nbsp;Shunpan Liang ,&nbsp;Fuzhi Zhang","doi":"10.1016/j.neunet.2026.108669","DOIUrl":"10.1016/j.neunet.2026.108669","url":null,"abstract":"<div><div>Bundle recommendation is designed to suggest a set of correlated items to a user in a holistic manner rather than recommending these items separately. Recent methods introduce <u>c</u>ontrastive <u>l</u>earning (CL) to refine the node representations learned from different graphs (generally termed the item and bundle views) for better recommendation performance. Unfortunately, these methods have two deficiencies. Firstly, few of them explicitly model the user-user and bundle-bundle relationships simultaneously from both the item and bundle views, leading to the underutilization of high-order relationships between users (bundles). Secondly, they use InfoNCE as the contrastive loss, which overlooks the graph structure as supervised signals in defining positive (negative) samples, resulting in anchor-like nodes being treated as negative samples. To tackle these deficiencies, an approach of <u>c</u>ross-view <u>c</u>ontrastive <u>r</u>epresentation <u>l</u>earning (CCRL) on meta-path induced graphs with node features is proposed for bundle recommendation. First, we introduce meta-path to model the user-user and bundle-bundle relationships as meta-path induced graphs with node features from both the item and bundle views. Second, we perform graph representation learning on the meta-path induced graphs with node features to procure the user (bundle) representations and introduce a contrastive loss that supports multiple positive samples to build a cross-view graph CL mechanism for refining the learned user (bundle) representations. Finally, the model is trained with a joint optimization objective. Experiments on the benchmark datasets manifest that our approach surpasses the baselines in bundle recommendation.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108669"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146114722","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}

{"title":"Improved exponential stability of time delay neural networks via separated-matrix-based integral inequalities","authors":"Yuanyuan Zhang,&nbsp;Xinzuo Ma,&nbsp;Seakweng Vong","doi":"10.1016/j.neunet.2026.108643","DOIUrl":"10.1016/j.neunet.2026.108643","url":null,"abstract":"<div><div>This paper studies the exponential stability of neural networks with time delays. A separated-matrix-based integral inequality is proposed to incorporate more delay information. It not only reflects the information of each component in the state-related vector but also considers the cross terms among the three components, significantly reducing the inherent conservativeness of traditional methods. By constructing a Lyapunov-Krasovskii functional with separation-matrix-based integral and a linear matrix inequality framework via quadratic negative definiteness, less conservative stability criteria are established. Two numerical examples demonstrate the method superiority in maximum allowable delay bounds and computational efficiency compared to existing approaches.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108643"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146114767","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}

{"title":"Adaptive sample repulsion against class-specific counterfactuals for explainable imbalanced classification","authors":"Yu Hao ,&nbsp;Xin Gao ,&nbsp;Xinping Diao ,&nbsp;Yuan Li ,&nbsp;Yukun Lin ,&nbsp;Tianyang Chen ,&nbsp;Qiangwei Li ,&nbsp;Jiawen Lu","doi":"10.1016/j.neunet.2026.108652","DOIUrl":"10.1016/j.neunet.2026.108652","url":null,"abstract":"<div><div>Enhancing model classification capability for samples within overlapping regions in complex feature spaces remains a key challenge in imbalanced classification research. Existing mainstream methods at the data-level and algorithm-level primarily rely on original sample distribution information to reduce overlap impact, without deeply modeling the causal relationship between features and labels. Furthermore, these approaches often overlook instance-level explanations that could guide deep discriminative information mining for samples of different classes in overlapping regions, thus the improvement on classification performance and model credibility may be constrained. This paper proposes an explainable imbalanced classification framework with adaptive sample repulsion against class-specific counterfactuals (CSCF-SR), forming a closed-loop between explanation generation and classification decisions by dynamically regulating the feature-space distribution through generated counterfactual samples. Two core phases are jointly optimized. (1) Counterfactual searching: a class-specific dual-actor architecture based on reinforcement learning decouples perturbation policy learning for majority and minority classes. A multi-step dynamic perturbation mechanism is designed to control counterfactual search behavior more precisely and smoothly, effectively generating reliable counterfactual samples. (2) Adaptive sample repulsion against counterfactuals: exploiting the inter-class discriminative information in displacement vectors between counterfactual and original samples, each original sample is adaptively perturbed along the direction opposite to its counterfactual. This fine-grained regulation gradually displaces samples from the overlapping region and clarifies class boundaries. Experiments on 50 imbalanced datasets demonstrate that CSCF-SR has a performance advantage over 27 typical imbalanced classification methods on both F1-score and G-mean, with more pronounced improvements on 25 datasets with severe class overlap.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108652"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146120771","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}

{"title":"Multi-step first: A lightweight deep reinforcement learning strategy for robust continuous control with partial observability","authors":"Lingheng Meng ,&nbsp;Rob Gorbet ,&nbsp;Michael Burke ,&nbsp;Dana Kulić","doi":"10.1016/j.neunet.2025.108521","DOIUrl":"10.1016/j.neunet.2025.108521","url":null,"abstract":"<div><div>Deep Reinforcement Learning (DRL) has made considerable advances in simulated and physical robot control tasks, especially when problems admit a fully observed Markov Decision Process (MDP) formulation. When observations only partially capture the underlying state, the problem becomes a Partially Observable MDP (POMDP), and performance rankings between algorithms can change. We empirically compare Proximal Policy Optimization (PPO), Twin Delayed Deep Deterministic Policy Gradient (TD3), and Soft Actor-Critic (SAC) on representative POMDP variants of continuous-control benchmarks. Contrary to widely reported MDP results where TD3 and SAC typically outperform PPO, we observe an inversion: PPO attains higher robustness under partial observability. We attribute this to the stabilizing effect of multi-step bootstrapping. Furthermore, incorporating multi-step targets into TD3 (MTD3) and SAC (MSAC) improves their robustness. These findings provide practical guidance for selecting and adapting DRL algorithms in partially observable settings without requiring new theoretical machinery.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108521"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146138146","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}