Neural Networks最新文献

{"title":"SHFormer: Dynamic spectral filtering convolutional neural network and high-pass kernel generation transformer for adaptive MRI reconstruction","authors":"Sriprabha Ramanarayanan ,&nbsp;Rahul G.S. ,&nbsp;Mohammad Al Fahim ,&nbsp;Keerthi Ram ,&nbsp;Ramesh Venkatesan ,&nbsp;Mohanasankar Sivaprakasam","doi":"10.1016/j.neunet.2025.107334","DOIUrl":"10.1016/j.neunet.2025.107334","url":null,"abstract":"<div><div>Attention Mechanism (AM) selectively focuses on essential information for imaging tasks and captures relationships between regions from distant pixel neighborhoods to compute feature representations. Accelerated magnetic resonance image (MRI) reconstruction can benefit from AM, as the imaging process involves acquiring Fourier domain measurements that influence the image representation in a non-local manner. However, AM-based models are more adept at capturing low-frequency information and have limited capacity in constructing high-frequency representations, restricting the models to smooth reconstruction. Secondly, AM-based models need mode-specific retraining for multimodal MRI data as their knowledge is restricted to local contextual variations within modes that might be inadequate to capture the diverse transferable features across heterogeneous data domains. To address these challenges, we propose a neuromodulation-based discriminative multi-spectral AM for scalable MRI reconstruction, that can (i) propagate the context-aware high-frequency details for high-quality image reconstruction, and (ii) capture features reusable to deviated unseen domains in multimodal MRI, to offer high practical value for the healthcare industry and researchers. The proposed network consists of a spectral filtering convolutional neural network to capture mode-specific transferable features to generalize to deviated MRI data domains and a dynamic high-pass kernel generation transformer that focuses on high-frequency details for improved reconstruction. We have evaluated our model on various aspects, such as comparative studies in supervised and self-supervised learning, diffusion model-based training, closed-set and open-set generalization under heterogeneous MRI data, and interpretation-based analysis. Our results show that the proposed method offers scalable and high-quality reconstruction with best improvement margins of <span><math><mo>∼</mo></math></span>1 dB in PSNR and <span><math><mo>∼</mo></math></span>0.01 in SSIM under unseen scenarios. Our code is available at <span><span>https://github.com/sriprabhar/SHFormer</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107334"},"PeriodicalIF":6.0,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143611705","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":"INN/ENNS/JNNS - Membership Applic. Form","authors":"","doi":"10.1016/S0893-6080(25)00239-4","DOIUrl":"10.1016/S0893-6080(25)00239-4","url":null,"abstract":"","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"186 ","pages":"Article 107360"},"PeriodicalIF":6.0,"publicationDate":"2025-03-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143576692","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Ternary spike-based neuromorphic signal processing system","authors":"Shuai Wang ,&nbsp;Dehao Zhang ,&nbsp;Ammar Belatreche ,&nbsp;Yichen Xiao ,&nbsp;Hongyu Qing ,&nbsp;Wenjie Wei ,&nbsp;Malu Zhang ,&nbsp;Yang Yang","doi":"10.1016/j.neunet.2025.107333","DOIUrl":"10.1016/j.neunet.2025.107333","url":null,"abstract":"<div><div>Deep Neural Networks (DNNs) have been successfully implemented across various signal processing fields, resulting in significant enhancements in performance. However, DNNs generally require substantial computational resources, leading to significant economic costs and posing challenges for their deployment on resource-constrained edge devices. In this study, we take advantage of spiking neural networks (SNNs) and quantization technologies to develop an energy-efficient and lightweight neuromorphic signal processing system. Our system is characterized by two principal innovations: a threshold-adaptive encoding (TAE) method and a quantized ternary SNN (QT-SNN). The TAE method can efficiently encode time-varying analog signals into sparse ternary spike trains, thereby reducing energy and memory demands for signal processing. QT-SNN, compatible with ternary spike trains from the TAE method, quantifies both membrane potentials and synaptic weights to reduce memory requirements while maintaining performance. Extensive experiments are conducted on two typical signal-processing tasks: speech and electroencephalogram recognition. The results demonstrate that our neuromorphic signal processing system achieves state-of-the-art (SOTA) performance with a 94% reduced memory requirement. Furthermore, through theoretical energy consumption analysis, our system shows <span><math><mrow><mn>7</mn><mo>.</mo><mn>5</mn><mo>×</mo></mrow></math></span> energy saving compared to other SNN works. The efficiency and efficacy of the proposed system highlight its potential as a promising avenue for energy-efficient signal processing.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107333"},"PeriodicalIF":6.0,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143601732","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":"Attribute-guided feature fusion network with knowledge-inspired attention mechanism for multi-source remote sensing classification","authors":"Xiao Pan,&nbsp;Changzhe Jiao,&nbsp;Bo Yang,&nbsp;Hao Zhu,&nbsp;Jinjian Wu","doi":"10.1016/j.neunet.2025.107332","DOIUrl":"10.1016/j.neunet.2025.107332","url":null,"abstract":"<div><div>Land use and land cover (LULC) classification is a popular research area in remote sensing. The information of single-modal data is insufficient for accurate classification, especially in complex scenes, while the complementarity of multi-modal data such as hyperspectral images (HSIs) and light detection and ranging (LiDAR) data could effectively improve classification performance. The attention mechanism has recently been widely used in multi-modal LULC classification methods to achieve better feature representation. However, the knowledge of data is insufficiently considered in these methods, such as spectral mixture in HSIs and inconsistent spatial scales of different categories in LiDAR data. Moreover, multi-modal features contain different physical attributes, HSI features can represent spectral information of several channels while LiDAR features focus on elevation information at the spatial dimension. Ignoring these attributes, feature fusion may introduce redundant information and effect detrimentally on classification. In this paper, we propose an attribute-guided feature fusion network with knowledge-inspired attention mechanisms, named AFNKA. Focusing on the spectral characteristics of HSI and elevation information of LiDAR data, we design the knowledge-inspired attention mechanism to explore enhanced features. Especially, a novel adaptive cosine estimator (ACE) based attention module is presented to learn features with more discriminability, which adequately utilizes the spatial–spectral correlation of HSI mixed pixels. In the fusion stage, two novel attribute-guided fusion modules are developed to selectively aggregate multi-modal features, which sufficiently exploit the correlations between the spatial–spectral property of HSI features and the spatial-elevation property of LiDAR features. Experimental results on several multi-source datasets quantitatively indicate that the proposed AFNKA significantly outperforms the state-of-the-art methods.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107332"},"PeriodicalIF":6.0,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143619139","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":"FedELR: When federated learning meets learning with noisy labels","authors":"Ruizhi Pu ,&nbsp;Lixing Yu ,&nbsp;Shaojie Zhan ,&nbsp;Gezheng Xu ,&nbsp;Fan Zhou ,&nbsp;Charles X. Ling ,&nbsp;Boyu Wang","doi":"10.1016/j.neunet.2025.107275","DOIUrl":"10.1016/j.neunet.2025.107275","url":null,"abstract":"<div><div>Existing research on federated learning (FL) usually assumes that training labels are of high quality for each client, which is impractical in many real-world scenarios (e.g., noisy labels by crowd-sourced annotations), leading to dramatic performance degradation. In this work, we investigate noisy FL through the lens of early-time training phenomenon (ETP). Specifically, a key finding of this paper is that the early training phase varies among different local clients due to the different noisy classes in each client. In addition, we show that such an inconsistency also exists between the local and global models. As a result, local clients would always begin to memorize noisy labels before the global model reaches the optimal, which inevitably leads to the degradation of the quality of service in real-world FL applications (e.g. tumor image classification among different hospitals). Our findings provide new insights into the learning dynamics and shed light on the essence cause of this degradation in noisy FL. To address this problem, we reveal a new principle for noisy FL: it is necessary to align the early training phases across local models. To this end, we propose FedELR, a simple yet effective framework that aims to force local models to stick to their early training phase via an early learning regularization (ELR), so that the learning dynamics of local models can be kept at the same pace. Moreover, this also leverages the ETP in local clients, leading each client to take more training steps in learning a more robust local model for optimal global aggregation. Extensive experiments on various real-world datasets also validate the effectiveness of our proposed methods.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107275"},"PeriodicalIF":6.0,"publicationDate":"2025-03-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143611712","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":"HCPI-HRL: Human Causal Perception and Inference-driven Hierarchical Reinforcement Learning","authors":"Bin Chen ,&nbsp;Zehong Cao ,&nbsp;Wolfgang Mayer ,&nbsp;Markus Stumptner ,&nbsp;Ryszard Kowalczyk","doi":"10.1016/j.neunet.2025.107318","DOIUrl":"10.1016/j.neunet.2025.107318","url":null,"abstract":"<div><div>The dependency on extensive expert knowledge for defining subgoals in hierarchical reinforcement learning (HRL) restricts the training efficiency and adaptability of HRL agents in complex, dynamic environments. Inspired by human-guided causal discovery skills, we proposed a novel method, Human Causal Perception and Inference-driven Hierarchical Reinforcement Learning (HCPI-HRL), designed to infer diverse, effective subgoal structures as intrinsic rewards and incorporate critical objects from dynamic environmental states using stable causal relationships. The HCPI-HRL method is supposed to guide an agent’s exploration direction and promote the reuse of learned subgoal structures across different tasks. Our designed HCPI-HRL comprises two levels: the top level operates as a meta controller, assigning subgoals discovered based on human-driven causal critical object perception and causal structure inference; the bottom level employs the Proximal Policy Optimisation (PPO) algorithm to accomplish the assigned subgoals. Experiments conducted across discrete and continuous control environments demonstrated that HCPI-HRL outperforms benchmark methods such as hierarchical and adjacency PPO in terms of training efficiency, exploration capability, and transferability. Our research extends the potential of HRL methods incorporating human-guided causal modelling to infer the effective relationships across subgoals, enhancing the agent’s capability to learn efficient policies in dynamic environments with sparse reward signals.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107318"},"PeriodicalIF":6.0,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143579133","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"ADAMT: Adaptive distributed multi-task learning for efficient image recognition in Mobile Ad-hoc Networks","authors":"Jia Zhao ,&nbsp;Wei Zhao ,&nbsp;Yunan Zhai ,&nbsp;Liyuan Zhang ,&nbsp;Yan Ding","doi":"10.1016/j.neunet.2025.107316","DOIUrl":"10.1016/j.neunet.2025.107316","url":null,"abstract":"<div><div>Distributed machine learning in mobile adhoc networks faces significant challenges due to the limited computational resources of devices, non-IID data distribution, and dynamic network topology. Existing approaches often rely on centralized coordination and stable network conditions, which may not be feasible in practice. To address these issues, we propose an adaptive distributed multi-task learning framework called ADAMT for efficient image recognition in resource-constrained mobile ad hoc networks. ADAMT introduces three key innovations: (1) a feature expansion mechanism that enhances the expressiveness of local models by leveraging task-specific information; (2) a deep hashing technique that enables efficient on-device retrieval and multi-task fusion; and (3) an adaptive communication strategy that dynamically adjusts the model updating process based on network conditions and node reliability. The proposed framework allows each device to perform personalized model training on its local dataset while collaboratively updating the shared parameters with neighboring nodes. Extensive experiments on the ImageNet dataset demonstrate the superiority of ADAMT over state-of-the-art methods. ADAMT achieves a top-1 accuracy of 0.867, outperforming existing distributed learning approaches. Moreover, ADAMT significantly reduces the communication overhead and accelerates the convergence speed by 2.69 times compared to traditional distributed SGD. The adaptive communication strategy effectively balances the trade-off between model performance and resource consumption, making ADAMT particularly suitable for resource-constrained environments. Our work sheds light on the design of efficient and robust distributed learning algorithms for mobile adhoc networks and paves the way for deploying advanced machine learning applications on edge devices.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107316"},"PeriodicalIF":6.0,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143592423","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":"Neural networks trained by weight permutation are universal approximators","authors":"Yongqiang Cai ,&nbsp;Gaohang Chen ,&nbsp;Zhonghua Qiao","doi":"10.1016/j.neunet.2025.107277","DOIUrl":"10.1016/j.neunet.2025.107277","url":null,"abstract":"<div><div>The universal approximation property is fundamental to the success of neural networks, and has traditionally been achieved by training networks without any constraints on their parameters. However, recent experimental research proposed a novel permutation-based training method, which exhibited a desired classification performance without modifying the exact weight values. In this paper, we provide a theoretical guarantee of this permutation training method by proving its ability to guide a ReLU network to approximate one-dimensional continuous functions. Our numerical results further validate this method’s efficiency in regression tasks with various initializations. The notable observations during weight permutation suggest that permutation training can provide an innovative tool for describing network learning behavior.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107277"},"PeriodicalIF":6.0,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143619141","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Balancing user preferences by social networks: A condition-guided social recommendation model for mitigating popularity bias","authors":"Xin He ,&nbsp;Wenqi Fan ,&nbsp;Ruobing Wang ,&nbsp;Yili Wang ,&nbsp;Ying Wang ,&nbsp;Shirui Pan ,&nbsp;Xin Wang","doi":"10.1016/j.neunet.2025.107317","DOIUrl":"10.1016/j.neunet.2025.107317","url":null,"abstract":"<div><div>Social recommendation models weave social interactions into their design to provide uniquely personalized recommendation results for users. However, social networks not only amplify the popularity bias in recommendation models, resulting in more frequent recommendation of hot items and fewer long-tail items, but also include a substantial amount of redundant information that is essentially meaningless for the model’s performance. Existing social recommendation models often integrate the entire social network directly, with little effort to filter or adjust social information to mitigate popularity bias introduced by the social network. In this paper, we propose a Condition-Guided Social Recommendation Model (named CGSoRec) to mitigate the model’s popularity bias by denoising the social network and adjusting the weights of user’s social preferences. More specifically, CGSoRec first includes a Condition-Guided Social Denoising Model (CSD) to remove redundant social relations in the social network for capturing users’ social preferences with items more precisely. Then, CGSoRec calculates users’ social preferences based on denoised social network and adjusts the weights in users’ social preferences to make them can counteract the popularity bias present in the recommendation model. At last, CGSoRec includes a Condition-Guided Diffusion Recommendation Model (CGD) to introduce the adjusted social preferences as conditions to control the recommendation results for a debiased direction. Comprehensive experiments on three real-world datasets demonstrate the effectiveness of our proposed method. The anonymous code is in: <span><span>https://anonymous.4open.science/r/CGSoRec-2B72</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107317"},"PeriodicalIF":6.0,"publicationDate":"2025-03-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143644930","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":"Deep Huber quantile regression networks","authors":"Hristos Tyralis ,&nbsp;Georgia Papacharalampous ,&nbsp;Nilay Dogulu ,&nbsp;Kwok P. Chun","doi":"10.1016/j.neunet.2025.107364","DOIUrl":"10.1016/j.neunet.2025.107364","url":null,"abstract":"<div><div>Typical machine learning regression applications aim to report the mean or the median of the predictive probability distribution, via training with a squared or an absolute error scoring function. The importance of issuing predictions of more functionals of the predictive probability distribution (quantiles and expectiles) has been recognized as a means to quantify the uncertainty of the prediction. In deep learning (DL) applications, that is possible through quantile and expectile regression neural networks (QRNN and ERNN respectively). Here we introduce deep Huber quantile regression networks (DHQRN) that nest QRNN and ERNN as edge cases. DHQRN can predict Huber quantiles, which are more general functionals in the sense that they nest quantiles and expectiles as limiting cases. The main idea is to train a DL algorithm with the Huber quantile scoring function, which is consistent for the Huber quantile functional. As a proof of concept, DHQRN are applied to predict house prices in Melbourne, Australia and Boston, United States (US). In this context, predictive performances of three DL architectures are discussed along with evidential interpretation of results from two economic case studies. Additional simulation experiments and applications to real-world case studies using open datasets demonstrate a satisfactory absolute performance of DHQRN.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"187 ","pages":"Article 107364"},"PeriodicalIF":6.0,"publicationDate":"2025-03-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143644929","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}