Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107468
Dan Zhang , Tong Zhang , Zhang Tao , C.L. Philip Chen
{"title":"Broad learning system based on fractional order optimization","authors":"Dan Zhang , Tong Zhang , Zhang Tao , C.L. Philip Chen","doi":"10.1016/j.neunet.2025.107468","DOIUrl":"10.1016/j.neunet.2025.107468","url":null,"abstract":"<div><div>Due to its efficient incremental learning performance, the broad learning system (BLS) has received widespread attention in the field of machine learning. Scholars have found in algorithm research that using the maximum correntropy criterion (MCC) can further improves the performance of broad learning in handling outliers. Recent studies have shown that differential equations can be used to represent the forward propagation of deep learning. The BLS based on MCC uses differentiation to optimize parameters, which indicates that differential methods can also be used for BLS optimization. But general methods use integer order differential equations, ignoring system information between integer orders. Due to the long-term memory property of fractional differential equations, this paper innovatively introduces fractional order optimization into the BLS, called FOBLS, to better enhance the data processing capability of the BLS. Firstly, a BLS is constructed using fractional order, incorporating long-term memory characteristics into the weight optimization process. In addition, constructing a dynamic incremental learning system based on fractional order further enhances the ability of network optimization. The experimental results demonstrate the excellent performance of the method proposed in this paper.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107468"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143864936","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}
Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107467
Meng Gao , Yutao Xie , Wei Chen , Feng Zhang , Fei Ding , Tengjiao Wang , Jiahui Yao , Jiabin Zheng , Kam-Fai Wong
{"title":"ReranKGC: A cooperative retrieve-and-rerank framework for multi-modal knowledge graph completion","authors":"Meng Gao , Yutao Xie , Wei Chen , Feng Zhang , Fei Ding , Tengjiao Wang , Jiahui Yao , Jiabin Zheng , Kam-Fai Wong","doi":"10.1016/j.neunet.2025.107467","DOIUrl":"10.1016/j.neunet.2025.107467","url":null,"abstract":"<div><div>Multi-modal knowledge graph completion (MMKGC) aims to predict missing links using entity’s multi-modal attributes. Embedding-based methods excel in leveraging structural knowledge, making them robust to entity ambiguity, yet their performance is constrained by the underutilization of multi-modal knowledge. Conversely, fine-tune-based (FT-based) approaches excel in extracting multi-modal knowledge but are hindered by ambiguity issues. To harness the complementary strengths of both methods for MMKGC, this paper introduces an ensemble framework <em>ReranKGC</em>, which decomposes KGC to a retrieve-and-rerank pipeline. The retriever employs embedding-based methods for initial retrieval. The re-ranker adopts our proposed KGC-CLIP, an FT-based method that utilizes CLIP to extract multi-modal knowledge from attributes for candidate re-ranking. By leveraging a more comprehensive knowledge source, the retriever generates a candidate pool containing entities not only semantically, but also structurally related to the query entity. Within this higher-quality candidate pool, the re-ranker can better discern candidates’ semantics to further refine the initial ranking, thereby enhancing precision. Through cooperation, each method maximizes its strengths while mitigating the weaknesses of others to a certain extent, leading to superior performance that surpasses individual capabilities. Extensive experiments conducted on link prediction tasks demonstrate that our framework ReranKGC consistently enhances baseline performance, outperforming state-of-the-art models.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107467"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143844164","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}
Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107450
Yiqian Luo , Qiurong Chen , Fali Li , Liang Yi , Peng Xu , Yangsong Zhang
{"title":"Hierarchical feature extraction on functional brain networks for autism spectrum disorder identification with resting-state fMRI data","authors":"Yiqian Luo , Qiurong Chen , Fali Li , Liang Yi , Peng Xu , Yangsong Zhang","doi":"10.1016/j.neunet.2025.107450","DOIUrl":"10.1016/j.neunet.2025.107450","url":null,"abstract":"<div><div>Autism Spectrum Disorder (ASD) is a pervasive developmental disorder of the central nervous system, primarily manifesting in childhood. It is characterized by atypical and repetitive behaviors. Conventional diagnostic methods mainly rely on questionnaire surveys and behavioral observations, which are prone to misdiagnosis due to their subjective nature. With advancements in medical imaging, MR imaging-based diagnostics have emerged as a more objective alternative. In this paper, we propose a Hierarchical Neural Network model for ASD identification, termed ASD-HNet, which hierarchically extracts features from functional brain networks based on resting-state functional magnetic resonance imaging (rs-fMRI) data. This hierarchical approach enhances the extraction of brain representations, improving diagnostic accuracy and aiding in the identification of brain regions associated with ASD. Specifically, features are extracted at three levels, i.e., the local region of interest (ROI) scale, the community scale, and the global representation scale. At the ROI scale, graph convolution is employed to transfer features between ROIs. At the community scale, functional gradients are introduced, and a K-Means clustering algorithm is applied to group ROIs with similar functional gradients into communities. Features from ROIs within the same community are then extracted to characterize the communities. At the global representation scale, we extract global features from the whole community-scale brain networks to represent the entire brain. We validate the effectiveness of the ASD-HNet model using the publicly available Autism Brain Imaging Data Exchange I (ABIDE-I) dataset, ADHD-200,dataset and ABIDE-II dataset. Extensive experimental results demonstrate that ASD-HNet outperforms existing baseline methods. The code is available at <span><span>https://github.com/LYQbyte/ASD-HNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107450"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143825552","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}
Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107432
Ye Tao , Jiawang Liu , Chaofeng Lu , Meng Liu , Xiugong Qin , Yunlong Tian , Yongjie Du
{"title":"CMDF-TTS: Text-to-speech method with limited target speaker corpus","authors":"Ye Tao , Jiawang Liu , Chaofeng Lu , Meng Liu , Xiugong Qin , Yunlong Tian , Yongjie Du","doi":"10.1016/j.neunet.2025.107432","DOIUrl":"10.1016/j.neunet.2025.107432","url":null,"abstract":"<div><div>While end-to-end Text-to-Speech (TTS) methods with limited target speaker corpus can generate high-quality speech, they often require a non-target speaker corpus (auxiliary corpus) which contains a substantial amount of <<em>text, speech</em>> pairs to train the model, significantly increasing training costs. In this work, we propose a fast and high-quality speech synthesis approach, requiring few target speaker recordings. Based on statistics, we analyzed the role of phonemes, function words, and utterance target domains in the corpus and proposed a Statistical-based Compression Auxiliary Corpus algorithm (SCAC). It significantly improves model training speed without a noticeable decrease in speech naturalness. Next, we use the compressed corpus to train the proposed non-autoregressive model CMDF-TTS, which uses a multi-level prosody modeling module to obtain more information and Denoising Diffusion Probabilistic Models (DDPMs) to generate mel-spectrograms. Besides, we fine-tune the model using the target speaker corpus to embed the speaker’s characteristics into the model and Conditional Variational Auto-Encoder Generative Adversarial Networks(CVAE-GAN) to enhance further the synthesized speech’s quality. Experimental results on multiple Mandarin and English corpus demonstrate that the CMDF-TTS model, enhanced by the SCAC algorithm, effectively balances training speed and synthesized speech quality. Overall, its performance surpasses that of state-of-the-art models.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107432"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143844168","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}
Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107464
Yuan Wang , Huaxin Pang , Ying Qin , Shikui Wei , Yao Zhao
{"title":"Adaptive estimation of instance-dependent noise transition matrix for learning with instance-dependent label noise","authors":"Yuan Wang , Huaxin Pang , Ying Qin , Shikui Wei , Yao Zhao","doi":"10.1016/j.neunet.2025.107464","DOIUrl":"10.1016/j.neunet.2025.107464","url":null,"abstract":"<div><div>Instance-dependent noise (IDN) widely exists in real-world datasets, seriously hindering the effective application of deep neural networks. In contrast to class-dependent noise, IDN is influenced not solely by the class but also by the intrinsic features of the instance. Current methods for addressing IDN typically involve estimating the instance-dependent noise transition matrix (IDNT). However, these approaches often either assume a specific form for the IDNT or rely on anchor points, which can result in significant estimation errors. To tackle this issue, we propose a method that makes no assumptions about the form of IDNT or the need for anchor points. Specifically, by computing similarity scores between each instance’s feature representation and the label representations, the instance-label confusion matrix (ILCM) captures the relationships between global instance features and different categories, providing valuable insights into the degree of noise. We then adaptively combine the noisy class posteriors from network predictions (for noisy data) or given labels (for clean data) with the ILCM, weighted by the degree of noise, to enhance the accuracy of IDNT estimation. Finally, the obtained IDNT adjusts the loss function, resulting in a more robust classifier. Comprehensive experiments comparing our method with state-of-the-art approaches on synthetic IDN datasets (F-MNIST, SVHN, CIFAR-10, CIFAR-100) and a real-world noisy dataset (Clothing1M) demonstrate the superiority and effectiveness of our approach.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107464"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143838013","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}
Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107461
Xindong Wang , Zidong Liu , Junghui Chen
{"title":"Collaborative twin actors framework using deep deterministic policy gradient for flexible batch processes","authors":"Xindong Wang , Zidong Liu , Junghui Chen","doi":"10.1016/j.neunet.2025.107461","DOIUrl":"10.1016/j.neunet.2025.107461","url":null,"abstract":"<div><div>Due to its inherent efficiency in the process industry for achieving desired products, batch processing is widely acknowledged for its repetitive nature. Batch-to-batch learning control has traditionally been esteemed as a robust strategy for batch process control. However, the presence of flexible operating conditions in practical batch systems often leads to a lack of prior learning information, hindering learning control from optimizing performance. This article presents a novel approach to flexible batch process control using deep reinforcement learning (DRL) with twin actors. Specifically, a collaborative twin-actor-based deep deterministic policy gradient (CTA-DDPG) method is proposed to generate control policies and ensure safe operation across varying trial lengths and initial conditions. This approach involves the sequential construction of two sets of actor–critic networks with a shared critic. The first set explores meta-policy during an offline stage, while the second set enhances control performance using a supplementary agent during an online stage. To ensure robust policy transfer and efficient learning, a policy integration mechanism and a spatial–temporal experience replay strategy are incorporated, facilitating transfer stability and learning efficiency. The performance of CTA-DDPG is evaluated using both numerical examples and nonlinear injection molding process for tracking control. The results demonstrate the effectiveness and superiority of the proposed method in achieving desired control outcomes.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107461"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143870674","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}
Neural NetworksPub Date : 2025-04-12DOI: 10.1016/j.neunet.2025.107466
Hengda Gao , Xiao-Wei Guo , Genglin Li , Chao Li , Canqun Yang
{"title":"GCPNet: An interpretable Generic Crystal Pattern graph neural Network for predicting material properties","authors":"Hengda Gao , Xiao-Wei Guo , Genglin Li , Chao Li , Canqun Yang","doi":"10.1016/j.neunet.2025.107466","DOIUrl":"10.1016/j.neunet.2025.107466","url":null,"abstract":"<div><div>To predict material properties from crystal structures, we introduce a simple yet flexible Generic Crystal Pattern graph neural Network (GCPNet), which is based on crystal pattern graphs and employs the Graph Convolutional Attention Operator (GCAO) along with a two-level update mechanism to extract key structural features from crystalline materials effectively. The GCPNet model complements the missing microstructure inputs of existing networks and leverages diverse information updating mechanisms, enabling the prediction of material properties with better precision over other networks on five public datasets. Further experiments show that our model is straightforward to use and robust in real-world applications. We also highlight the good interpretability of GCPNet, using local contributions from our model to increase the search efficiency for the high-throughput perovskite screening by 32%. Taken together, our findings show that the GCPNet model offers an effective solution to facilitate the screening and discovery of ideal crystals and is an efficient alternative to existing neural networks in material property prediction.The implementation code can be found at <span><span>https://github.com/feiji110/GCPNet</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107466"},"PeriodicalIF":6.0,"publicationDate":"2025-04-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143833248","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}
Neural NetworksPub Date : 2025-04-11DOI: 10.1016/j.neunet.2025.107451
Zhibin Shi , Zhenghong Lin , Weihong Lin , Shiping Wang
{"title":"Exploring unified cross-view hypergraph generation for multi-view semi-supervised classification","authors":"Zhibin Shi , Zhenghong Lin , Weihong Lin , Shiping Wang","doi":"10.1016/j.neunet.2025.107451","DOIUrl":"10.1016/j.neunet.2025.107451","url":null,"abstract":"<div><div>Graph structure is widely used in the field of multi-view learning. Hypergraph which is a kind of extension of graph can capture the higher-order relationships of nodes in a better way. However, most existing hypergraph-based models are based on the assumption that hypergraph structures are readily available, which is untenable in most cases. In order to alleviate this problem, we propose the learnable unified hypergraph dynamic system framework, a novel approach in unified cross-view hypergraph structure generation tailored for multi-view semi-supervised classification. Specifically, we introduce four strategies for unified cross-view hypergraph generation and propose a mechanism for generating learnable unified cross-view hypergraph. Furthermore, we utilize a dynamic diffusion model to dynamically learn unified hypergraph structure which can achieve better performance in multi-view semi-supervised classification tasks. Extensive experimental results on various real datasets show that the proposed method outperforms other state-of-the-art multi-view algorithms.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107451"},"PeriodicalIF":6.0,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143833246","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}
Neural NetworksPub Date : 2025-04-11DOI: 10.1016/j.neunet.2025.107463
Junyi Wu , Yan Huang , Min Gao , Yuzhen Niu , Yuzhong Chen , Qiang Wu
{"title":"High-order diversity feature learning for pedestrian attribute recognition","authors":"Junyi Wu , Yan Huang , Min Gao , Yuzhen Niu , Yuzhong Chen , Qiang Wu","doi":"10.1016/j.neunet.2025.107463","DOIUrl":"10.1016/j.neunet.2025.107463","url":null,"abstract":"<div><div>Pedestrian attribute recognition (PAR) involves accurately identifying multiple attributes present in pedestrian images. There are two main approaches for PAR: part-based method and attention-based method. The former relies on existing segmentation or region detection methods to localize body parts and learn corresponding attribute-specific feature from the corresponding regions, where the performance heavily depends on the accuracy of body region localization. The latter adopts the embedded attention modules or transformer attention to exploit detailed feature. However, it can focus on certain body regions but often provide coarse attention, failing to capture fine-grained details, the learned feature may also be interfered with by irrelevant information. Meanwhile, these methods overlook the global contextual information. This work argues for replacing coarse attention with detailed attention and integrating it with global contextual feature from ViT to jointly represent attribute-specific regions. To tackle this issue, we propose a High-order Diversity Feature Learning (HDFL) method for PAR based on ViT. We utilize a polynomial predictor to design an Attribute-specific Detailed Feature Exploration (ADFE) module, which can construct the high-order statistics and gain more fine-grained feature. Our ADFE module is a parameter-friendly method that provides flexibility in deciding its utilization during the inference phase. A Soft-redundancy Perception Loss (SPLoss) is proposed to adaptively measure the redundancy between feature of different orders, which can promote diverse characterization of features. Experiments on several PAR datasets show that our method achieves a new state-of-the-art (SOTA) performance. On the most challenging PA100K dataset, our method outperforms previous SOTA by 1.69% and achieves the highest mA of 84.92%.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107463"},"PeriodicalIF":6.0,"publicationDate":"2025-04-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143850863","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}
Neural NetworksPub Date : 2025-04-10DOI: 10.1016/j.neunet.2025.107449
Jiahao Qin , Feng Liu , Lu Zong
{"title":"BC-PMJRS: A Brain Computing-inspired Predefined Multimodal Joint Representation Spaces for enhanced cross-modal learning","authors":"Jiahao Qin , Feng Liu , Lu Zong","doi":"10.1016/j.neunet.2025.107449","DOIUrl":"10.1016/j.neunet.2025.107449","url":null,"abstract":"<div><div>Multimodal learning faces two key challenges: effectively fusing complex information from different modalities, and designing efficient mechanisms for cross-modal interactions. Inspired by neural plasticity and information processing principles in the human brain, this paper proposes BC-PMJRS, a Brain Computing-inspired Predefined Multimodal Joint Representation Spaces method to enhance cross-modal learning. The method learns the joint representation space through two complementary optimization objectives: (1) minimizing mutual information between representations of different modalities to reduce redundancy and (2) maximizing mutual information between joint representations and sentiment labels to improve task-specific discrimination. These objectives are balanced dynamically using an adaptive optimization strategy inspired by long-term potentiation (LTP) and long-term depression (LTD) mechanisms. Furthermore, we significantly reduce the computational complexity of modal interactions by leveraging a global–local cross-modal interaction mechanism, analogous to selective attention in the brain. Experimental results on the IEMOCAP, MOSI, and MOSEI datasets demonstrate that BC-PMJRS outperforms state-of-the-art models in both complete and incomplete modality settings, achieving up to a 1.9% improvement in weighted-F1 on IEMOCAP, a 2.8% gain in 7-class accuracy on MOSI, and a 2.9% increase in 7-class accuracy on MOSEI. These substantial improvements across multiple datasets demonstrate that incorporating brain-inspired mechanisms, particularly the dynamic balance of information redundancy and task relevance through neural plasticity principles, effectively enhances multimodal learning. This work bridges neuroscience principles with multimodal machine learning, offering new insights for developing more effective and biologically plausible models.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"188 ","pages":"Article 107449"},"PeriodicalIF":6.0,"publicationDate":"2025-04-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143821249","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}