Neural Networks最新文献

{"title":"TENet: Targetness entanglement incorporating with multi-scale pooling and mutually-guided fusion for RGB-E object tracking.","authors":"Pengcheng Shao, Tianyang Xu, Zhangyong Tang, Linze Li, Xiao-Jun Wu, Josef Kittler","doi":"10.1016/j.neunet.2024.106948","DOIUrl":"https://doi.org/10.1016/j.neunet.2024.106948","url":null,"abstract":"<p><p>There is currently strong interest in improving visual object tracking by augmenting the RGB modality with the output of a visual event camera that is particularly informative about the scene motion. However, existing approaches perform event feature extraction for RGB-E tracking using traditional appearance models, which have been optimised for RGB only tracking, without adapting it for the intrinsic characteristics of the event data. To address this problem, we propose an Event backbone (Pooler), designed to obtain a high-quality feature representation that is cognisant of the innate characteristics of the event data, namely its sparsity. In particular, Multi-Scale Pooling is introduced to capture all the motion feature trends within event data through the utilisation of diverse pooling kernel sizes. The association between the derived RGB and event representations is established by an innovative module performing adaptive Mutually Guided Fusion (MGF). Extensive experimental results show that our method significantly outperforms state-of-the-art trackers on two widely used RGB-E tracking datasets, including VisEvent and COESOT, where the precision and success rates on COESOT are improved by 4.9% and 5.2%, respectively. Our code will be available at https://github.com/SSSpc333/TENet.</p>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"106948"},"PeriodicalIF":6.0,"publicationDate":"2024-11-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142808460","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-loss, feature fusion and improved top-two-voting ensemble for facial expression recognition in the wild","authors":"Guangyao Zhou ,&nbsp;Yuanlun Xie ,&nbsp;Yiqin Fu ,&nbsp;Zhaokun Wang","doi":"10.1016/j.neunet.2024.106937","DOIUrl":"10.1016/j.neunet.2024.106937","url":null,"abstract":"<div><div>Facial expression recognition (FER) in the wild is a challenging pattern recognition task affected by the images’ low quality and has attracted broad interest in computer vision. Existing FER methods failed to obtain sufficient accuracy to support the practical applications, especially in scenarios with low fault tolerance, which limits the adaptability of FER. Targeting exploring the possibility of further improving the accuracy of FER in the wild, this paper proposes a novel single model named R18+FAML and an ensemble model named R18+FAML-FGA-T2V, which applies intra-feature fusion within a single network, feature fusion among multiple networks, and the ensemble decision strategy. Based on the backbone of ResNet18 (R18), R18+FAML combines internal feature fusion and three attention blocks, as well as uses multiple loss functions (FAML) to improve the diversity of the feature extraction. To effectively integrate feature extractors from multiple networks, we propose feature fusion among networks based on the genetic algorithm (FGA). Comprehensively considering and utilizing more classification information, we propose an ensemble strategy, i.e., the improved top-two-voting (T2V) of multiple networks with the same structure. Combining the above strategies, R18+FAML-FGA-T2V can focus on the main expression-aware areas by integrating interest areas of multiple networks. From experiments on three challenging FER datasets in the wild including RAF-DB, AffectNet-8 and AffectNet-7, our single model R18+FAML and ensemble model R18+FAML-FGA-T2V achieve the accuracies of <span><math><mrow><mfenced><mrow><mn>90</mn><mo>.</mo><mn>32</mn><mo>,</mo><mn>62</mn><mo>.</mo><mn>17</mn><mo>,</mo><mn>65</mn><mo>.</mo><mn>83</mn></mrow></mfenced><mtext>%</mtext></mrow></math></span> and <span><math><mrow><mfenced><mrow><mn>91</mn><mo>.</mo><mn>59</mn><mo>,</mo><mn>63</mn><mo>.</mo><mn>27</mn><mo>,</mo><mn>66</mn><mo>.</mo><mn>63</mn></mrow></mfenced><mtext>%</mtext></mrow></math></span> respectively, both achieving the state-of-the-art results.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"Article 106937"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744928","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":"Two-step graph propagation for incomplete multi-view clustering","authors":"Xiao Zhang ,&nbsp;Xinyu Pu ,&nbsp;Hangjun Che ,&nbsp;Cheng Liu ,&nbsp;Jun Qin","doi":"10.1016/j.neunet.2024.106944","DOIUrl":"10.1016/j.neunet.2024.106944","url":null,"abstract":"<div><div>Incomplete multi-view clustering addresses scenarios where data completeness cannot be guaranteed, diverging from traditional methods that assume fully observed features. Existing approaches often overlook high-order correlations present in multiple similarity graphs, and suffer from inefficiencies due to iterative optimization procedures. To overcome these limitations, we propose a graph-based model leveraging graph propagation to effectively handle incomplete data. The proposed method translates incomplete instances into incomplete graphs, and infers missing entries through a graph propagation strategy, ensuring the inferred data is meaningful and contextually relevant. Specifically, a self-guided graph is constructed to capture global relationships, while partial graphs represent view-specific similarities. The self-guided graph is first completed through self-guided graph propagation, which subsequently aids in the propagation of the partial graphs. The key contribution of graph propagation is to propagate information from complete data to incomplete data. Furthermore, the high-order correlation across multiple views is captured by low-rank tensor learning. To enhance computational efficiency, the optimization procedure is decoupled and implemented in a stepwise manner, eliminating the need for iterative updates. Extensive experiments validate the robustness of the proposed method, demonstrating superior performance compared to state-of-the-art methods, even when all instances are incomplete.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"Article 106944"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142757043","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":"Revisiting the problem of learning long-term dependencies in recurrent neural networks.","authors":"Liam Johnston, Vivak Patel, Yumian Cui, Prasanna Balaprakash","doi":"10.1016/j.neunet.2024.106887","DOIUrl":"https://doi.org/10.1016/j.neunet.2024.106887","url":null,"abstract":"<p><p>Recurrent neural networks (RNNs) are an important class of models for learning sequential behavior. However, training RNNs to learn long-term dependencies is a tremendously difficult task, and this difficulty is widely attributed to the vanishing and exploding gradient (VEG) problem. Since it was first characterized 30 years ago, the belief that if VEG occurs during optimization then RNNs learn long-term dependencies poorly has become a central tenet in the RNN literature and has been steadily cited as motivation for a wide variety of research advancements. In this work, we revisit and interrogate this belief using a large factorial experiment where more than 40,000 RNNs were trained, and provide evidence contradicting this belief. Motivated by these findings, we re-examine the original discussion that analyzed latching behavior in RNNs by way of hyperbolic attractors, and ultimately demonstrate that these dynamics do not fully capture the learned characteristics of RNNs. Our findings suggest that these models are fully capable of learning dynamics that do not correspond to hyperbolic attractors, and that the choice of hyper-parameters, namely learning rate, has a substantial impact on the likelihood of whether an RNN will be able to learn long-term dependencies.</p>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"106887"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142787487","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":"Spike-VisNet: A novel framework for visual recognition with FocusLayer-STDP learning","authors":"Ying Liu ,&nbsp;Xiaoling Luo ,&nbsp;Ya Zhang ,&nbsp;Yun Zhang ,&nbsp;Wei Zhang ,&nbsp;Hong Qu","doi":"10.1016/j.neunet.2024.106918","DOIUrl":"10.1016/j.neunet.2024.106918","url":null,"abstract":"<div><div>Current vision-inspired spiking neural networks (SNNs) face key challenges due to their model structures typically focusing on single mechanisms and neglecting the integration of multiple biological features. These limitations, coupled with limited synaptic plasticity, hinder their ability to implement biologically realistic visual processing. To address these issues, we propose Spike-VisNet, a novel retina-inspired framework designed to enhance visual recognition capabilities. This framework simulates both the functional and layered structure of the retina. To further enhance this architecture, we integrate the FocusLayer-STDP learning rule, allowing Spike-VisNet to dynamically adjust synaptic weights in response to varying visual stimuli. This rule combines channel attention, inhibition mechanisms, and competitive mechanisms with spike-timing-dependent plasticity (STDP), significantly improving synaptic adaptability and visual recognition performance. Comprehensive evaluations on benchmark datasets demonstrate that Spike-VisNet outperforms other STDP-based SNNs, achieving precision scores of 98.6% on MNIST, 93.29% on ETH-80, and 86.27% on CIFAR-10. These results highlight its effectiveness and robustness, showcasing Spike-VisNet’s potential to simulate human visual processing and its applicability to complex real-world visual challenges.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"182 ","pages":"Article 106918"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142743443","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":"D4A: An efficient and effective defense across agnostic adversarial attacks","authors":"Xianxian Li ,&nbsp;Zeming Gan ,&nbsp;Yan Bai ,&nbsp;Linlin Su ,&nbsp;De Li ,&nbsp;Jinyan Wang","doi":"10.1016/j.neunet.2024.106938","DOIUrl":"10.1016/j.neunet.2024.106938","url":null,"abstract":"<div><div>Recent studies show that Graph Neural Networks (GNNs) are vulnerable to structure adversarial attacks, which draws attention to adversarial defenses in graph data. Previous defenses designed heuristic defense strategies for specific attacks or graph properties, and are no longer sufficiently robust across all these attacks. To address this problem, we discuss the abnormal behaviors of GNNs in structure perturbations from a posterior distribution perspective. We suggest that the structural vulnerability of GNNs stems from their dependence on local graph smoothing, which can also lead to <em>unfitting</em> — a first-found phenomenon specific to the graph domain. We demonstrate that abnormal behaviors, except for unfitting, can attribute to a posterior distribution shift. To intrinsically prevent the occurrence of abnormal behaviors, we first propose smooth-less message passing to enhance the tolerance with respect to structure perturbations, while significantly mitigating the unfitting. We also propose the distribution shift constraint to restrict other abnormal behaviors of our model. Our approach is evaluated on six different datasets across over four kinds of attacks and compared to 11 representative baselines. The experimental results show that our method improves the defense performance across various attacks, and provides a great trade-off between accuracy and adversarial robustness.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"Article 106938"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744921","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":"Arithmetic abilities of SNP systems with astrocytes producing calcium","authors":"Bogdan Aman ,&nbsp;Gabriel Ciobanu","doi":"10.1016/j.neunet.2024.106913","DOIUrl":"10.1016/j.neunet.2024.106913","url":null,"abstract":"<div><div>Are the membrane systems able of performing arithmetic operations? In the last dozen years, there were published several implementations of the arithmetic operations based on membrane systems by using all available topologies (cell-like, tissue-like, or neural-like). In particular, the spiking neural P systems perform arithmetic operations by using the numbers represented in binary base. In this paper, we consider numbers represented in unary base (to each number n corresponds an object with multiplicity n), and we propose two encodings for the main arithmetic operations (addition, subtraction, multiplication and division) between numbers given in unary base: (i) for each pair of input values generate an instance of a spiking neural P system with astrocytes producing calcium with rules based on these values; (ii) generate a spiking neural P system with astrocytes producing calcium that does not depend on these values. While the second approach is commonly used in membrane computing to construct only a system for each operation, the first approach is interesting because each system is uniquely constructed based on a pair of input values , and so it performs faster the desired arithmetic operation. The main advantage (with respect to other attempts) of using any of these two approaches to perform arithmetic operations consists in the reduced size of created systems (number of locations and used rules). Additionally, we extend a semantic interpreter (in Haskell) for spiking neural P systems to test all the encodings of the arithmetic operations presented in this paper.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"Article 106913"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142744922","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":"BPEN: Brain Posterior Evidential Network for trustworthy brain imaging analysis.","authors":"Kai Ye, Haoteng Tang, Siyuan Dai, Igor Fortel, Paul M Thompson, R Scott Mackin, Alex Leow, Heng Huang, Liang Zhan","doi":"10.1016/j.neunet.2024.106943","DOIUrl":"https://doi.org/10.1016/j.neunet.2024.106943","url":null,"abstract":"<p><p>The application of deep learning techniques to analyze brain functional magnetic resonance imaging (fMRI) data has led to significant advancements in identifying prospective biomarkers associated with various clinical phenotypes and neurological conditions. Despite these achievements, the aspect of prediction uncertainty has been relatively underexplored in brain fMRI data analysis. Accurate uncertainty estimation is essential for trustworthy learning, given the challenges associated with brain fMRI data acquisition and the potential diagnostic implications for patients. To address this gap, we introduce a novel posterior evidential network, named the Brain Posterior Evidential Network (BPEN), designed to capture both aleatoric and epistemic uncertainty in the analysis of brain fMRI data. We conducted comprehensive experiments using data from the Alzheimer's Disease Neuroimaging Initiative (ADNI) and ADNI-depression (ADNI-D) cohorts, focusing on predictions for mild cognitive impairment (MCI) and depression across various diagnostic groups. Our experiments not only unequivocally demonstrate the superior predictive performance of our BPEN model compared to existing state-of-the-art methods but also underscore the importance of uncertainty estimation in predictive models.</p>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"106943"},"PeriodicalIF":6.0,"publicationDate":"2024-11-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142808452","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":"RC-DETR: Improving DETRs in crowded pedestrian detection via rank-based contrastive learning","authors":"Feng Gao,&nbsp;Jiaxu Leng,&nbsp;Ji Gan,&nbsp;Xinbo Gao","doi":"10.1016/j.neunet.2024.106911","DOIUrl":"10.1016/j.neunet.2024.106911","url":null,"abstract":"<div><div>The variants of DEtection TRansformer (DETRs) have achieved impressive performance in general object detection. However, they suffer notable performance degradation in scenarios involving crowded pedestrian detection. This decline primarily occurs during the training phase, where DETRs are constrained solely by pedestrian labels. This limitation leads to the production of indistinguishable image features between visually similar pedestrians and background elements, resulting in incorrect detections. To address this issue, this paper introduces a rank-based contrastive learning method, which constructs an additional and specific constraint for each indistinguishable training sample to produce distinguishable image features. Unlike previous methods that rely solely on pedestrian labels to achieve a consistent confidence score, our approach relies on multiple constraints and aims to ensure the correct rank of detection results, with confidence scores of pedestrians consistently surpassing those of background elements. Specifically, we first filter out some training samples that could interfere with our delineation of indistinguishable and distinguishable training samples. Then, based on the confidence score rank, we divide the rest of the training samples into distinguishable positive and negative training samples and indistinguishable positive and negative training samples. Finally, we combine these training samples into multiple positive and negative pairs and utilize these sample pairs to train DETRs via contrastive learning. Our method can be plugged into any DETRs and does not increase any overhead on inference. Extensive experiments on three DETRs show that our method achieves superior performance. Especially on the Crowdhuman dataset, our method achieved the state-of-the-art 38.9% MR.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"182 ","pages":"Article 106911"},"PeriodicalIF":6.0,"publicationDate":"2024-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142743444","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":"Optimization control for mean square synchronization of stochastic semi-Markov jump neural networks with non-fragile hidden information and actuator saturation.","authors":"Zou Yang, Jun Wang, Kaibo Shi, Xiao Cai, Sheng Han","doi":"10.1016/j.neunet.2024.106942","DOIUrl":"https://doi.org/10.1016/j.neunet.2024.106942","url":null,"abstract":"<p><p>This paper studies the asynchronous output feedback control and H_∞ synchronization problems for a class of continuous-time stochastic hidden semi-Markov jump neural networks (SMJNNs) affected by actuator saturation. Initially, a novel neural networks (NNs) model is constructed, incorporating semi-Markov process (SMP), hidden information, and Brownian motion to accurately simulate the complexity and uncertainty of real-world environments. Secondly, acknowledging system mode mismatches and the need for robust anti-interference capabilities, a non-fragile controller based on hidden information is proposed. The designed controller effectively mitigates the impact of uncertainties enhancing system reliability. Furthermore, sufficient conditions for stochastic mean square synchronization (MSS) within the domain of attraction are provided, and optimal control is achieved through the construction of a Lyapunov function based on SMP. Finally, the feasibility of the proposed method is verified through numerical examples.</p>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"183 ","pages":"106942"},"PeriodicalIF":6.0,"publicationDate":"2024-11-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142792280","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}