Neural Networks最新文献_第2页

UniSymNet: A Unified Symbolic Network with Sparse Encoding and Bi-level Optimization 具有稀疏编码和双级优化的统一符号网络。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-22 DOI: 10.1016/j.neunet.2026.108615

Xinxin Li , Juan Zhang , Da Li , Xingyu Liu , Jin Xu , Junping Yin

Automatically discovering mathematical expressions is a challenging issue to precisely depict natural phenomena, in which Symbolic Regression (SR) is one of the most widely utilized techniques. Mainstream SR algorithms target on searching for the optimal symbolic tree, but the increasing complexity of the tree structure often limits their performance. Inspired by neural networks, symbolic networks have emerged as a promising new paradigm. However, existing symbolic networks still face certain challenges: binary nonlinear operators { × , ÷} cannot be naturally extended to multivariate, training with fixed architecture often leads to higher complexity and overfitting. In this work, we propose a Unified Symbolic Network that unifies nonlinear binary operators into nested unary operators, thereby transforming them into multivariate operators. The capability of the proposed UniSymNet is deduced from rigorous theoretical proof, resulting in lower complexity and stronger expressivity. Unlike the conventional neural network training, we design a bi-level optimization framework: the outer level pre-trains a Transformer with sparse label encoding scheme to guide UniSymNet structure selection, while the inner level employs objective-specific strategies to optimize network parameters. This allows for flexible adaptation of UniSymNet structures to different data, leading to reduced expression complexity. The UniSymNet is evaluated on low-dimensional Standard Benchmarks and high-dimensional SRBench, and shows excellent symbolic solution rate, high fitting accuracy, and relatively low expression complexity.

自动发现数学表达式是精确描述自然现象的一个具有挑战性的问题，其中符号回归（SR）是应用最广泛的技术之一。主流SR算法的目标是寻找最优的符号树，但树结构的复杂性往往限制了它们的性能。受神经网络的启发，符号网络已经成为一种很有前途的新范式。然而，现有的符号网络仍然面临着一定的挑战：二元非线性算子{ × ，÷}不能自然地扩展到多元，固定架构的训练往往会导致更高的复杂性和过拟合。在这项工作中，我们提出了一个统一的符号网络，将非线性二进制算子统一为嵌套的一元算子，从而将它们转换为多元算子。通过严格的理论论证，推导出了该网络的性能，具有较低的复杂度和较强的表达能力。与传统的神经网络训练不同，我们设计了一个双层优化框架：外层预训练一个具有稀疏标签编码方案的Transformer来指导UniSymNet的结构选择，而内层采用特定目标的策略来优化网络参数。这允许灵活地调整UniSymNet结构以适应不同的数据，从而降低表达式的复杂性。在低维的Standard benchmark和高维的SRBench上对UniSymNet进行了评估，结果表明该算法具有优异的符号解算率、较高的拟合精度和相对较低的表达式复杂度。

{"title":"UniSymNet: A Unified Symbolic Network with Sparse Encoding and Bi-level Optimization","authors":"Xinxin Li , Juan Zhang , Da Li , Xingyu Liu , Jin Xu , Junping Yin","doi":"10.1016/j.neunet.2026.108615","DOIUrl":"10.1016/j.neunet.2026.108615","url":null,"abstract":"<div><div>Automatically discovering mathematical expressions is a challenging issue to precisely depict natural phenomena, in which Symbolic Regression (SR) is one of the most widely utilized techniques. Mainstream SR algorithms target on searching for the optimal symbolic tree, but the increasing complexity of the tree structure often limits their performance. Inspired by neural networks, symbolic networks have emerged as a promising new paradigm. However, existing symbolic networks still face certain challenges: binary nonlinear operators { × , ÷} cannot be naturally extended to multivariate, training with fixed architecture often leads to higher complexity and overfitting. In this work, we propose a <strong>Uni</strong>fied <strong>Sym</strong>bolic <strong>Net</strong>work that unifies nonlinear binary operators into nested unary operators, thereby transforming them into multivariate operators. The capability of the proposed UniSymNet is deduced from rigorous theoretical proof, resulting in lower complexity and stronger expressivity. Unlike the conventional neural network training, we design a bi-level optimization framework: the outer level pre-trains a Transformer with sparse label encoding scheme to guide UniSymNet structure selection, while the inner level employs objective-specific strategies to optimize network parameters. This allows for flexible adaptation of UniSymNet structures to different data, leading to reduced expression complexity. The UniSymNet is evaluated on low-dimensional Standard Benchmarks and high-dimensional SRBench, and shows excellent symbolic solution rate, high fitting accuracy, and relatively low expression complexity.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108615"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146108216","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Emotion-Aware multimodal deepfake detection 情感感知多模态深度假检测。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-31 DOI: 10.1016/j.neunet.2026.108675

Teng Zhang , Gen Li , Yanhui Xiao , Huawei Tian , Yun Cao

With the continuous advancement of Deepfake techniques, traditional unimodal detection methods struggle to address the challenges posed by multimodal manipulations. Most existing approaches rely on large-scale training data, which limits their generalization to unseen identities or different manipulation types in few-shot settings. In this paper, we propose an emotion-aware multimodal Deepfake detection method that exploits emotion signals for forgery detection. Specifically, we design an emotion embedding extractor (Emoencoder) to capture emotion representations within modalities. Then, we employ Emotion-Aware Contrastive Learning and Cross-Modal Contrastive Learning to capture cross-modal inconsistencies and enhance modality feature extraction. Furthermore, we propose a Text-Guided Semantic Fusion module, where the text modality serves as a semantic anchor to guide audio-visual feature interactions for multimodal feature fusion. To validate our approach under data-limited conditions and unseen identities, we employ a cross-identity few-shot training strategy on benchmark datasets. Experimental results demonstrate that our method outperforms SOTAs and demonstrates superior generalization to both unseen identities and manipulation types.

随着Deepfake技术的不断进步，传统的单峰检测方法难以应对多峰操作带来的挑战。大多数现有的方法依赖于大规模的训练数据，这限制了它们在少数镜头设置中对看不见的身份或不同操作类型的泛化。在本文中，我们提出了一种利用情感信号进行伪造检测的情感感知多模态深度伪造检测方法。具体来说，我们设计了一个情感嵌入提取器（Emoencoder）来捕获模态中的情感表征。然后，我们采用情绪感知对比学习和跨模态对比学习来捕捉跨模态不一致性，增强模态特征提取。此外，我们提出了一个文本引导语义融合模块，其中文本情态作为语义锚来指导多模态特征融合的视听特征交互。为了在数据有限的条件和不可见的身份下验证我们的方法，我们在基准数据集上采用了交叉身份的少量训练策略。实验结果表明，我们的方法优于sota，并且对看不见的身份和操作类型都有更好的泛化。

{"title":"Emotion-Aware multimodal deepfake detection","authors":"Teng Zhang , Gen Li , Yanhui Xiao , Huawei Tian , Yun Cao","doi":"10.1016/j.neunet.2026.108675","DOIUrl":"10.1016/j.neunet.2026.108675","url":null,"abstract":"<div><div>With the continuous advancement of Deepfake techniques, traditional unimodal detection methods struggle to address the challenges posed by multimodal manipulations. Most existing approaches rely on large-scale training data, which limits their generalization to unseen identities or different manipulation types in few-shot settings. In this paper, we propose an emotion-aware multimodal Deepfake detection method that exploits emotion signals for forgery detection. Specifically, we design an emotion embedding extractor (Emoencoder) to capture emotion representations within modalities. Then, we employ Emotion-Aware Contrastive Learning and Cross-Modal Contrastive Learning to capture cross-modal inconsistencies and enhance modality feature extraction. Furthermore, we propose a Text-Guided Semantic Fusion module, where the text modality serves as a semantic anchor to guide audio-visual feature interactions for multimodal feature fusion. To validate our approach under data-limited conditions and unseen identities, we employ a cross-identity few-shot training strategy on benchmark datasets. Experimental results demonstrate that our method outperforms SOTAs and demonstrates superior generalization to both unseen identities and manipulation types.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108675"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146133401","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Observer-based prescribed-time optimal neural consensus control for six-rotor UAVs: A novel actor-critic reinforcement learning strategy 基于观测器的六旋翼无人机规定时间最优神经共识控制：一种新的actor-critic强化学习策略。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-02-01 DOI: 10.1016/j.neunet.2026.108644

Yue Zhou , Liang Cao , Yan Lei , Hongru Ren

Six-rotor unmanned aerial vehicles (UAVs) offer significant potential, but still encounter persistent challenges in achieving efficient allocation of limited resources in dynamic and complex environments. Consequently, this paper explores the prescribed-time observer-based optimal consensus control problem for six-rotor UAVs with unified prescribed performance. A practical prescribed-time optimal control scheme is constructed through embedding the prescribed-time control method with a simplified reinforcement learning framework to realize the efficient resource allocation. Leveraging a prescribed-time adjustment function, the novel updating laws for actor and critic neural networks are developed, which guarantee that six-rotor UAVs reach a desired steady state within prescribed time. Moreover, an improved distributed prescribed-time observer is established, ensuring that each follower is able to precisely estimate the velocity and position information of the leader within prescribed time. Then, a series of nonlinear transformations and mappings is proposed, which cannot only satisfy diverse performance requirements under a unified control framework through only adjusting the design parameters a priori but also improve the user-friendliness of implementation and control design. Significantly, the global performance requirement simplifies verification process of initial constraints in traditional performance control methods. Furthermore, an adaptive prescribed-time filter is introduced to address the complexity explosion issue of the backstepping method on six-rotor UAVs, while ensuring the filter error converges within prescribed time. Eventually, simulation results confirm the effectiveness of the designed method.

六旋翼无人机（uav）具有巨大的潜力，但在动态和复杂环境中实现有限资源的有效分配仍然面临着持续的挑战。因此，本文研究了具有统一规定性能的六旋翼无人机的基于规定时间观测器的最优一致控制问题。通过将规定时间控制方法嵌入简化的强化学习框架，构造了一个实用的规定时间最优控制方案，实现了资源的有效配置。利用规定时间的调节函数，提出了新的行动者和评论家神经网络的更新规律，保证了六旋翼无人机在规定时间内达到期望的稳态。建立了改进的分布式规定时间观测器，保证了每个follower都能在规定时间内准确估计leader的速度和位置信息。在此基础上，提出了一系列的非线性变换和映射，不仅可以通过先验地调整设计参数来满足统一控制框架下不同的性能要求，而且提高了实现和控制设计的用户友好性。重要的是，全局性能需求简化了传统性能控制方法中初始约束的验证过程。在保证滤波误差在规定时间内收敛的前提下，引入了自适应规定时间滤波器，解决了六旋翼无人机反步法的复杂度爆炸问题。最后，仿真结果验证了所设计方法的有效性。

{"title":"Observer-based prescribed-time optimal neural consensus control for six-rotor UAVs: A novel actor-critic reinforcement learning strategy","authors":"Yue Zhou , Liang Cao , Yan Lei , Hongru Ren","doi":"10.1016/j.neunet.2026.108644","DOIUrl":"10.1016/j.neunet.2026.108644","url":null,"abstract":"<div><div>Six-rotor unmanned aerial vehicles (UAVs) offer significant potential, but still encounter persistent challenges in achieving efficient allocation of limited resources in dynamic and complex environments. Consequently, this paper explores the prescribed-time observer-based optimal consensus control problem for six-rotor UAVs with unified prescribed performance. A practical prescribed-time optimal control scheme is constructed through embedding the prescribed-time control method with a simplified reinforcement learning framework to realize the efficient resource allocation. Leveraging a prescribed-time adjustment function, the novel updating laws for actor and critic neural networks are developed, which guarantee that six-rotor UAVs reach a desired steady state within prescribed time. Moreover, an improved distributed prescribed-time observer is established, ensuring that each follower is able to precisely estimate the velocity and position information of the leader within prescribed time. Then, a series of nonlinear transformations and mappings is proposed, which cannot only satisfy diverse performance requirements under a unified control framework through only adjusting the design parameters a priori but also improve the user-friendliness of implementation and control design. Significantly, the global performance requirement simplifies verification process of initial constraints in traditional performance control methods. Furthermore, an adaptive prescribed-time filter is introduced to address the complexity explosion issue of the backstepping method on six-rotor UAVs, while ensuring the filter error converges within prescribed time. Eventually, simulation results confirm the effectiveness of the designed method.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108644"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146138110","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Differentially private data augmentation via LLM generation with discriminative and distribution-aligned filtering 通过具有判别和分布对齐过滤的LLM生成差分私有数据增强。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-29 DOI: 10.1016/j.neunet.2026.108668

Yiping Song , Juhua Zhang , Zhiliang Tian , Taishu Sheng , Yuxin Yang , Minlie Huang , Xinwang Liu , Dongsheng Li

Data augmentation (DA) is a widely adopted approach for mitigating data insufficiency. Conducting DA in private domains requires privacy-preserving text generation, including anonymization or perturbation applied to sensitive textual data. The above methods lack formal protection guarantees. Existing Differential Privacy (DP) learning methods provide theoretical guarantees by adding calibrated noise to models or outputs. However, the large output space and model scales in text generation require substantial noise, which severely degrades synthesis quality. In this paper, we transfer DP-based synthetic sample generation to DP-based sample discrimination. Specifically, we propose a DP-based DA framework with a large language model (LLM) and a DP-based discriminator for private-domain text generation. Our key idea is to (1) leverage LLMs to generate large-scale high-quality samples, (2) select synthesized samples fitting the private domain, and (3) align the label distribution with the private domain. To achieve this, we use knowledge distillation to construct a DP-based discriminator: teacher models, accessing private data, guide a student model to select samples under calibrated noise. A DP-based tutor further constrains the label distribution of synthesized samples with a low privacy budget. We theoretically analyze the privacy guarantees and empirically validate our method on three medical text classification datasets, showing that our DP-synthesized samples significantly outperform state-of-the-art DP fine-tuning baselines in utility.

数据增强（DA）是一种广泛采用的缓解数据不足的方法。在私有领域进行数据分析需要保护隐私的文本生成，包括对敏感文本数据进行匿名化或扰动处理。上述方法缺乏正式的保护保证。现有的差分隐私（DP）学习方法通过在模型或输出中添加校准噪声来提供理论保证。然而，文本生成中较大的输出空间和模型尺度需要大量的噪声，这严重降低了合成质量。在本文中，我们将基于dp的合成样本生成转移到基于dp的样本判别。具体来说，我们提出了一个基于dp的数据分析框架，该框架具有大型语言模型（LLM）和一个用于私有领域文本生成的基于dp的鉴别器。我们的关键思想是：(1)利用llm来生成大规模的高质量样本，(2)选择拟合私有域的合成样本，(3)将标签分布与私有域对齐。为了实现这一点，我们使用知识蒸馏来构建一个基于dp的鉴别器：教师模型，访问私人数据，引导学生模型在校准噪声下选择样本。基于dp的导师进一步约束了合成样本的标签分布，隐私预算较低。我们从理论上分析了隐私保证，并在三个医学文本分类数据集上实证验证了我们的方法，表明我们的DP合成样本在实用性上明显优于最先进的DP微调基线。

{"title":"Differentially private data augmentation via LLM generation with discriminative and distribution-aligned filtering","authors":"Yiping Song , Juhua Zhang , Zhiliang Tian , Taishu Sheng , Yuxin Yang , Minlie Huang , Xinwang Liu , Dongsheng Li","doi":"10.1016/j.neunet.2026.108668","DOIUrl":"10.1016/j.neunet.2026.108668","url":null,"abstract":"<div><div>Data augmentation (DA) is a widely adopted approach for mitigating data insufficiency. Conducting DA in private domains requires privacy-preserving text generation, including anonymization or perturbation applied to sensitive textual data. The above methods lack formal protection guarantees. Existing Differential Privacy (DP) learning methods provide theoretical guarantees by adding calibrated noise to models or outputs. However, the large output space and model scales in text generation require substantial noise, which severely degrades synthesis quality. In this paper, we transfer DP-based synthetic sample generation to DP-based sample discrimination. Specifically, we propose a DP-based DA framework with a large language model (LLM) and a DP-based discriminator for private-domain text generation. Our key idea is to (1) leverage LLMs to generate large-scale high-quality samples, (2) select synthesized samples fitting the private domain, and (3) align the label distribution with the private domain. To achieve this, we use knowledge distillation to construct a DP-based discriminator: teacher models, accessing private data, guide a student model to select samples under calibrated noise. A DP-based tutor further constrains the label distribution of synthesized samples with a low privacy budget. We theoretically analyze the privacy guarantees and empirically validate our method on three medical text classification datasets, showing that our DP-synthesized samples significantly outperform state-of-the-art DP fine-tuning baselines in utility.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108668"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146138106","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Lifelong knowledge graph embedding via diffusion model 基于扩散模型的终身知识图谱嵌入

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-21 DOI: 10.1016/j.neunet.2026.108630

Deyu Chen , Caicai Guo , Qiyuan Li , Jinguang Gu , Meiyi Xie , Hong Zhu

Lifelong knowledge graph embedding (KGE) methods aim to learn new knowledge continuously while retaining old knowledge. This line of work has received much attention for its potential to enable knowledge retention and transfer and to reduce training costs under knowledge graphs’ growing scale and flexibility. However, embedding space drift under different contexts is a crucial reason for catastrophic forgetting and inefficient learning of new facts, and existing work ignores this perspective. In order to address the above issues, we proposed a novel lifelong KGE framework that considers learning new facts and preserving old facts in a unified perspective. We propose a diffusion-based embedding method that captures the contextual variation of entity representations and obtains transferable embeddings. In order to handle the drift of the embedding space and balance the learning efficiency, we adopt a reconstruction and generation strategy based on contrastive learning. To avoid catastrophic forgetting and maintain the stability of the embedding distribution, we proposed an effective distribution regularization method. We conduct extensive experiments on seven benchmark datasets with different construction strategies and incremental speed. Experimental results show that our proposed framework outperforms existing lifelong KGE methods.

终身知识图嵌入（KGE）方法的目的是在保留旧知识的同时不断学习新知识。在知识图的规模和灵活性不断增长的情况下，这方面的工作因其实现知识保留和转移以及降低培训成本的潜力而受到广泛关注。然而，在不同情境下嵌入空间漂移是灾难性遗忘和新事实学习效率低下的重要原因，现有的研究忽视了这一观点。为了解决上述问题，我们提出了一种新的终身KGE框架，该框架从统一的角度考虑了学习新事实和保留旧事实。我们提出了一种基于扩散的嵌入方法，该方法捕获实体表示的上下文变化并获得可转移的嵌入。为了处理嵌入空间的漂移和平衡学习效率，我们采用了一种基于对比学习的重构生成策略。为了避免灾难性遗忘和保持嵌入分布的稳定性，提出了一种有效的分布正则化方法。我们在7个基准数据集上采用不同的构建策略和增量速度进行了广泛的实验。实验结果表明，我们提出的框架优于现有的终身KGE方法。

{"title":"Lifelong knowledge graph embedding via diffusion model","authors":"Deyu Chen , Caicai Guo , Qiyuan Li , Jinguang Gu , Meiyi Xie , Hong Zhu","doi":"10.1016/j.neunet.2026.108630","DOIUrl":"10.1016/j.neunet.2026.108630","url":null,"abstract":"<div><div>Lifelong knowledge graph embedding (KGE) methods aim to learn new knowledge continuously while retaining old knowledge. This line of work has received much attention for its potential to enable knowledge retention and transfer and to reduce training costs under knowledge graphs’ growing scale and flexibility. However, embedding space drift under different contexts is a crucial reason for catastrophic forgetting and inefficient learning of new facts, and existing work ignores this perspective. In order to address the above issues, we proposed a novel lifelong KGE framework that considers learning new facts and preserving old facts in a unified perspective. We propose a diffusion-based embedding method that captures the contextual variation of entity representations and obtains transferable embeddings. In order to handle the drift of the embedding space and balance the learning efficiency, we adopt a reconstruction and generation strategy based on contrastive learning. To avoid catastrophic forgetting and maintain the stability of the embedding distribution, we proposed an effective distribution regularization method. We conduct extensive experiments on seven benchmark datasets with different construction strategies and incremental speed. Experimental results show that our proposed framework outperforms existing lifelong KGE methods.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108630"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146081676","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Learnable dendrite neural P systems and applications in survival prediction of glioblastoma patients 可学习树突神经P系统及其在胶质母细胞瘤患者生存预测中的应用。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-29 DOI: 10.1016/j.neunet.2026.108660

Xiu Yin , Xiyu Liu , Shulei Chang , Bosheng Song , Guanzhong Gong , Jiaxing Yin , Dengwang Li , Jie Xue

Current neural-like P systems use “point neurons” as the computing entities, and the computations in these neurons are simplified, ignoring the fact that, in organisms, subcellular compartments (such as neuronal dendrites) can also perform operations as independent computing units in addition to computing at the individual neuron level. The nervous system has a strong ability for optimization learning. Therefore, we propose learnable dendrite neural P (LDNP) systems with new plasticity rules, in which the dendrite structure and learning function can be adaptively changed when solving different application problems. Specifically, the dendrites of neurons are designed as dendritic trees composed of multiple dendritic branches, each of which serves as an independent computing unit. The multilevel complex topological structure of dendrites provides powerful computing capabilities for neurons. A model for predicting the overall survival of glioblastoma (GBM) patients was developed based on LDNP systems and validated on the GBM cohort from the Cancer Genome Atlas. Compared with thirteen state-of-the-art methods, the LDNP system achieves the best performance.

目前的类神经P系统使用“点神经元”作为计算实体，并且这些神经元中的计算被简化，忽略了这样一个事实，即在生物体中，亚细胞区室（如神经元树突）除了在单个神经元水平上进行计算外，还可以作为独立的计算单元执行操作。神经系统具有很强的优化学习能力。因此，我们提出了具有新的可塑性规则的可学习树突神经P （LDNP）系统，该系统的树突结构和学习功能可以在解决不同的应用问题时自适应改变。具体来说，神经元的树突被设计成由多个树突分支组成的树突树，每个树突分支作为一个独立的计算单元。树突的多层次复杂拓扑结构为神经元提供了强大的计算能力。基于LDNP系统开发了一个预测胶质母细胞瘤（GBM）患者总生存期的模型，并在来自癌症基因组图谱的GBM队列上进行了验证。与13种最先进的方法相比，LDNP系统的性能最好。

{"title":"Learnable dendrite neural P systems and applications in survival prediction of glioblastoma patients","authors":"Xiu Yin , Xiyu Liu , Shulei Chang , Bosheng Song , Guanzhong Gong , Jiaxing Yin , Dengwang Li , Jie Xue","doi":"10.1016/j.neunet.2026.108660","DOIUrl":"10.1016/j.neunet.2026.108660","url":null,"abstract":"<div><div>Current neural-like P systems use “point neurons” as the computing entities, and the computations in these neurons are simplified, ignoring the fact that, in organisms, subcellular compartments (such as neuronal dendrites) can also perform operations as independent computing units in addition to computing at the individual neuron level. The nervous system has a strong ability for optimization learning. Therefore, we propose learnable dendrite neural P (LDNP) systems with new plasticity rules, in which the dendrite structure and learning function can be adaptively changed when solving different application problems. Specifically, the dendrites of neurons are designed as dendritic trees composed of multiple dendritic branches, each of which serves as an independent computing unit. The multilevel complex topological structure of dendrites provides powerful computing capabilities for neurons. A model for predicting the overall survival of glioblastoma (GBM) patients was developed based on LDNP systems and validated on the GBM cohort from the Cancer Genome Atlas. Compared with thirteen state-of-the-art methods, the LDNP system achieves the best performance.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108660"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146127098","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A cortico-cerebellar neural model for task control under incomplete instructions 不完全指令下任务控制的皮质-小脑神经模型。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-27 DOI: 10.1016/j.neunet.2026.108648

Lanyun Cui , Ying Yu , Qingyun Wang , Guanrong Chen

Cerebellar-inspired motor control systems have been widely explored in robotics to achieve biologically plausible movement generation. However, most existing models rely heavily on high-dimensional instruction inputs during training, diverging from the input-efficient control observed in biological systems. In humans, effective motor learning often based on sparse or incomplete external feedback. It is possibly attributed to the interaction between multiple brain regions, especially the cortex and the cerebellum. In this study, we present a hierarchical cortico-cerebellar neural network model that investigates the neural mechanisms enabling motor control under incomplete or low-dimensional instructions. The evaluation results, measured by two complementary levels of evaluation metrics, demonstrate that the cortico-cerebellar model reduces dependency on external instruction without compromising trajectory smoothness. The model features a division of roles: the cortical network handles high-level action selection, while the cerebellar network executes motor commands by torque control, directly operating on a planar arm. Additionally, the cortex exhibits enhanced exploration indirectly driven by the stochastic characteristics of cerebellar torque control. Our results show that cortico-cerebellar coordination can facilitate robust and flexible control even with sparse instruction signals, suggesting a potential mechanism by which biological systems achieve efficient behavior under informational constraints.

以小脑为灵感的运动控制系统在机器人技术中得到了广泛的探索，以实现生物学上合理的运动生成。然而，大多数现有模型在训练过程中严重依赖高维指令输入，偏离了在生物系统中观察到的输入效率控制。在人类中，有效的运动学习通常基于稀疏或不完整的外部反馈。这可能归因于大脑多个区域，特别是皮层和小脑之间的相互作用。在这项研究中，我们提出了一个分层皮质-小脑神经网络模型，该模型研究了在不完整或低维指令下实现运动控制的神经机制。通过两个互补级别的评估指标测量的评估结果表明，皮质-小脑模型在不影响轨迹平滑的情况下减少了对外部指令的依赖。该模型具有角色划分的特点：皮质网络处理高级动作选择，而小脑网络通过扭矩控制执行运动命令，直接在平面手臂上操作。此外，小脑转矩控制的随机特性间接驱动了皮层探索能力的增强。我们的研究结果表明，即使在稀疏的指令信号下，皮质-小脑协调也可以促进鲁棒和灵活的控制，这表明生物系统在信息约束下实现有效行为的潜在机制。

{"title":"A cortico-cerebellar neural model for task control under incomplete instructions","authors":"Lanyun Cui , Ying Yu , Qingyun Wang , Guanrong Chen","doi":"10.1016/j.neunet.2026.108648","DOIUrl":"10.1016/j.neunet.2026.108648","url":null,"abstract":"<div><div>Cerebellar-inspired motor control systems have been widely explored in robotics to achieve biologically plausible movement generation. However, most existing models rely heavily on high-dimensional instruction inputs during training, diverging from the input-efficient control observed in biological systems. In humans, effective motor learning often based on sparse or incomplete external feedback. It is possibly attributed to the interaction between multiple brain regions, especially the cortex and the cerebellum. In this study, we present a hierarchical cortico-cerebellar neural network model that investigates the neural mechanisms enabling motor control under incomplete or low-dimensional instructions. The evaluation results, measured by two complementary levels of evaluation metrics, demonstrate that the cortico-cerebellar model reduces dependency on external instruction without compromising trajectory smoothness. The model features a division of roles: the cortical network handles high-level action selection, while the cerebellar network executes motor commands by torque control, directly operating on a planar arm. Additionally, the cortex exhibits enhanced exploration indirectly driven by the stochastic characteristics of cerebellar torque control. Our results show that cortico-cerebellar coordination can facilitate robust and flexible control even with sparse instruction signals, suggesting a potential mechanism by which biological systems achieve efficient behavior under informational constraints.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108648"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146133416","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Sonar-neus:voxel-based efficient neural implicit surface reconstruction for forward-looking sonar sonar - news：基于体素的高效神经隐式前视声纳表面重建。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-01-29 DOI: 10.1016/j.neunet.2026.108664

Shiji Qiu , Zuoqi Hu , Tiange Zhang , Zhi Liu , Junyu Dong , Qing Cai

Dense 3D reconstruction using forward-looking sonar (FLS) is essential for ocean exploration. Recent advancements in FLS-based 3D reconstruction using neural radiance fields have emerged, demonstrating promising results. However, their excessively slow reconstruction speed significantly impacts their application in real-world scenarios, primarily due to two reasons: (1) the reliance on MLPs for scene representation leads to slow training, often requiring several hours for reconstruction; and (2) the uniform sampling strategy along the elevation arc is inefficient, greatly hindering both training speed and reconstruction quality. To address these challenges, we propose a voxel-based efficient neural implicit surface reconstruction approach using FLS, featuring three key innovations: 1) Replacing MLPs with voxel grids for scene representation, utilizing a signed distance function (SDF) voxel grid to model geometry and a feature voxel grid to capture appearance. 2) Introducing a hierarchical sampling strategy along the elevation arc to improve sampling efficiency. 3) Applying SDF Gaussian convolution to the SDF voxel grid, effectively reducing noise and surface roughness. Extensive experiments demonstrate that our method significantly outperforms existing unsupervised dense FLS reconstruction techniques. Notably, our approach achieves the same reconstruction quality in just 10 minutes of training that previously required 4 hours with state-of-the-art methods, while also delivering superior results. We will open-source our code upon paper acceptance.

利用前视声呐（FLS）进行密集三维重建是海洋探测的关键。利用神经辐射场进行基于fls的三维重建的最新进展已经出现，显示出有希望的结果。然而，它们过于缓慢的重建速度显著影响了它们在现实场景中的应用，主要有两个原因：(1)依赖mlp进行场景表示导致训练缓慢，通常需要几个小时的重建；(2)沿仰角弧均匀采样策略效率低下，极大地影响了训练速度和重建质量。为了解决这些挑战，我们提出了一种基于体素的高效神经隐式表面重建方法，该方法使用FLS，具有三个关键创新：1)用体素网格代替mlp用于场景表示，利用符号距离函数（SDF）体素网格来建模几何形状，利用特征体素网格来捕获外观。2)引入沿高程弧线分层采样策略，提高采样效率。3)对SDF体素网格进行SDF高斯卷积，有效降低噪声和表面粗糙度。大量的实验表明，我们的方法明显优于现有的无监督密集FLS重建技术。值得注意的是，我们的方法只需10分钟的训练就可以实现相同的重建质量，而以前使用最先进的方法需要4个小时，同时也提供了卓越的结果。我们将在论文被接受后开放我们的代码。

{"title":"Sonar-neus:voxel-based efficient neural implicit surface reconstruction for forward-looking sonar","authors":"Shiji Qiu , Zuoqi Hu , Tiange Zhang , Zhi Liu , Junyu Dong , Qing Cai","doi":"10.1016/j.neunet.2026.108664","DOIUrl":"10.1016/j.neunet.2026.108664","url":null,"abstract":"<div><div>Dense 3D reconstruction using forward-looking sonar (FLS) is essential for ocean exploration. Recent advancements in FLS-based 3D reconstruction using neural radiance fields have emerged, demonstrating promising results. However, their excessively slow reconstruction speed significantly impacts their application in real-world scenarios, primarily due to two reasons: (1) the reliance on MLPs for scene representation leads to slow training, often requiring several hours for reconstruction; and (2) the uniform sampling strategy along the elevation arc is inefficient, greatly hindering both training speed and reconstruction quality. To address these challenges, we propose a voxel-based efficient neural implicit surface reconstruction approach using FLS, featuring three key innovations: 1) Replacing MLPs with voxel grids for scene representation, utilizing a signed distance function (SDF) voxel grid to model geometry and a feature voxel grid to capture appearance. 2) Introducing a hierarchical sampling strategy along the elevation arc to improve sampling efficiency. 3) Applying SDF Gaussian convolution to the SDF voxel grid, effectively reducing noise and surface roughness. Extensive experiments demonstrate that our method significantly outperforms existing unsupervised dense FLS reconstruction techniques. Notably, our approach achieves the same reconstruction quality in just 10 minutes of training that previously required 4 hours with state-of-the-art methods, while also delivering superior results. We will open-source our code upon paper acceptance.</div></div>","PeriodicalId":49763,"journal":{"name":"Neural Networks","volume":"199 ","pages":"Article 108664"},"PeriodicalIF":6.3,"publicationDate":"2026-07-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146101006","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Structure-missing graph-level clustering network 缺少结构的图级聚类网络。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-02-02 DOI: 10.1016/j.neunet.2026.108682

Tianyu Hu , Renda Han , Liu Mao , Jing Chen , Xia Xie

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 https://github.com/MrHuSN/SMGCN.

图级聚类旨在基于共享的结构特征或语义相似性将图分组为不同的聚类。然而，现有的图级聚类方法通常假设输入图结构是完整的，忽略了现实场景中普遍存在的关系缺失问题。这些未建模的缺失关系会导致图表示学习过程中结构信息失真的积累，显著降低聚类性能。为此，我们提出了一种新的方法——结构缺失图级聚类网络（SMGCN），该方法包括结构增强模块LR-SEA、锚定位机制和联合对比优化。具体而言，我们首先基于低秩矩阵补全输出增广图，使用匈牙利算法进行聚类匹配以获得锚点，然后强制相同的聚类图收敛到嵌入空间中相应的锚点。根据我们的研究，这是第一次提出具有缺失关系的图级聚类任务，并通过在五个基准数据集上的实验证明了我们的方法与现有方法的优越性。我们的源代码可在https://github.com/MrHuSN/SMGCN上获得。

{"title":"Structure-missing graph-level clustering network","authors":"Tianyu Hu , Renda Han , Liu Mao , Jing Chen , 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}

引用次数: 0

Loyalty-SMOTE: Data synthesis algorithm for effective imbalanced data classification loyty - smote：一种有效的不平衡数据分类的数据综合算法。

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

Neural Networks

Pub Date : 2026-07-01 Epub Date: 2026-02-03 DOI: 10.1016/j.neunet.2026.108677

Shengquan Hu , Junfei Li , Zefeng Li , Zihao Zhang , Yan Feng , K. L Eddie Law

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 Loyalty-SMOTE algorithm. We introduce the concept of Loyalty 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 Attraction 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.

不平衡的数据集在训练机器学习模型中总是存在问题，因此分类器经常难以达到令人满意的性能。已经开发了许多方法来解决数据不平衡问题。其中，一些数据级方法在相邻的少数类样本之间进行线性插值来生成新的数据点，而另一些方法则侧重于对特定类的边界样本进行过采样。然而，许多方法没有考虑到涉及噪声敏感性的情况。在本文中，我们提出了一种新的数据级方法，称为忠诚- smote算法。我们引入了忠诚度的概念来识别数据集中的噪声和边界。在识别出潜在的噪声数据点后，采用SMOTE （Synthetic Minority Oversampling Technique）算法对少数类边界数据进行过采样。然后，进行基于忠诚度的去噪处理，得到一个平衡的数据集。为了扩展我们的设计，引入了吸引力的概念来推广多类问题的去噪技术。在我们的研究中，使用SVM（支持向量机）分类器作为我们的基础学习器，并进行了大量的实验来评估和比较不同的算法。我们的研究结果表明，在二进制和多类UCI数据集上，loyal - smote在多个指标上都取得了卓越的性能。在30个二元数据集中，f1得分最高的数据集有26个（87%），AUROC得分最高的数据集有29个（97%），recall得分最高的数据集有26个（87%），G-mean得分最高的数据集有27个（90%）。对于5个多类数据集，我们的设计得分分别为0.8317、0.6153、0.8537和0.6717。

{"title":"Loyalty-SMOTE: Data synthesis algorithm for effective imbalanced data classification","authors":"Shengquan Hu , Junfei Li , Zefeng Li , Zihao Zhang , Yan Feng , 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}

引用次数: 0