Query：
学者姓名：孙艳丰
Refining：
Year
Type
Indexed by
Source
Complex
Co-Author
Language
Clean All
Abstract ：
基于层次多粒度交互图卷积网络的长文档分类方法及装置，在控制模型计算复杂度的情况下，能够构建网络以刻画长文档完备的层次结构化信息，以及进行图间信息交互。方法包括：(1)获得长文档层次化多粒度表示；(2)执行多层层次叠加的段落图卷积、句子图卷积和单词图卷积，以及相应的图间交互；(3)为了融合不同粒度不同尺度的语义信息，使用最大池化分别聚合段落图的终层输出，以及句子图和单词图每一层的输出。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 胡永利 , 刘腾飞 , 孙艳丰 et al. 基于层次多粒度交互图卷积网络的长文档分类方法及装置 : CN202310316635.1[P]. | 2023-03-24 . |
| MLA | 胡永利 et al. "基于层次多粒度交互图卷积网络的长文档分类方法及装置" : CN202310316635.1. | 2023-03-24 . |
| APA | 胡永利 , 刘腾飞 , 孙艳丰 , 尹宝才 . 基于层次多粒度交互图卷积网络的长文档分类方法及装置 : CN202310316635.1. | 2023-03-24 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
基于跨模态多粒度交互融合的长文档分类方法及装置，能够有效弥补现有方法对视觉信息的忽视，通过引入特征偏移网络在不同粒度实现跨模态的交互和融合，控制计算复杂度，达到分类准确率和分类效率的平衡。方法包括：(1)输入一个长文档中对应的文本序列，以及对应的单张或多张图片；(2)分别通过预训练编码器BERT和VGG‑16提取对应模态的多粒度特征表示；(3)使用多模态协同池化模块，在视觉信息和文本信息的协同引导下池化细粒度文本特征；(4)使用跨模态特征偏移网络，分别在4个不同的粒度组合下实现跨模态特征的交互和融合；(5)使用特征聚合网络实现多空间特征的融合，并获得最终的长文档分类结果。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 胡永利 , 刘腾飞 , 孙艳丰 et al. 基于跨模态多粒度交互融合的长文档分类方法及装置 : CN202310301100.7[P]. | 2023-03-24 . |
| MLA | 胡永利 et al. "基于跨模态多粒度交互融合的长文档分类方法及装置" : CN202310301100.7. | 2023-03-24 . |
| APA | 胡永利 , 刘腾飞 , 孙艳丰 , 尹宝才 . 基于跨模态多粒度交互融合的长文档分类方法及装置 : CN202310301100.7. | 2023-03-24 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
Graph Neural Networks (GNNs) have emerged as a crucial deep learning framework for graph-structured data. However, existing GNNs suffer from the scalability limitation, which hinders their practical implementation in industrial settings. Many scalable GNNs have been proposed to address this limitation. However, they have been proven to act as low-pass graph filters, which discard the valuable middle-and high-frequency information. This paper proposes a novel graph neural network named Adaptive Filtering Graph Neural Networks (AFGNN), which can capture all frequency information on large-scale graphs. AFGNN consists of two stages. The first stage utilizes low-, middle-, and high-pass graph filters to extract comprehensive frequency information without introducing additional parameters. This computation is a one-time task and is pre-computed before training, ensuring its scalability. The second stage incorporates a node-level attention -based feature combination, enabling the generation of customized graph filters for each node, contrary to existing spectral GNNs that employ uniform graph filters for the entire graph. AFGNN is suitable for mini-batch training, and can enhance scalability and efficiently capture all frequency information from large-scale graphs. We evaluate AFGNN by comparing its ability to capture all frequency information with spectral GNNs, and its scalability with scalable GNNs. Experimental results illustrate that AFGNN surpasses both scalable GNNs and spectral GNNs, highlighting its superiority.
Keyword ：
Spectral graph neural networks Spectral graph neural networks Scalable graph neural networks Scalable graph neural networks Large-scale graphs Large-scale graphs Graph signal filtering Graph signal filtering
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Zhang, Qi , Li, Jinghua , Sun, Yanfeng et al. Beyond low-pass filtering on large-scale graphs via Adaptive Filtering Graph Neural Networks [J]. | NEURAL NETWORKS , 2023 , 169 : 1-10 . |
| MLA | Zhang, Qi et al. "Beyond low-pass filtering on large-scale graphs via Adaptive Filtering Graph Neural Networks" . | NEURAL NETWORKS 169 (2023) : 1-10 . |
| APA | Zhang, Qi , Li, Jinghua , Sun, Yanfeng , Wang, Shaofan , Gao, Junbin , Yin, Baocai . Beyond low-pass filtering on large-scale graphs via Adaptive Filtering Graph Neural Networks . | NEURAL NETWORKS , 2023 , 169 , 1-10 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
Graph convolutional networks (GCNs) have become a popular tool for learning unstructured graph data due to their powerful learning ability. Many researchers have been interested in fusing topological structures and node features to extract the correlation information for classification tasks. However, it is inadequate to integrate the embedding from topology and feature spaces to gain the most correlated information. At the same time, most GCN-based methods assume that the topology graph or feature graph is compatible with the properties of GCNs, but this is usually not satisfied since meaningless, missing, or even unreal edges are very common in actual graphs. To obtain a more robust and accurate graph structure, we intend to construct an adaptive graph with topology and feature graphs. We propose Multi-graph Fusion Graph Convolutional Networks with pseudo-label supervision (MFGCN), which learn a connected embedding by fusing the multi-graphs and node features. We can obtain the final node embedding for semi-supervised node classification by propagating node features over multi-graphs. Furthermore, to alleviate the problem of labels missing in semi-supervised classification, a pseudo-label generation mechanism is proposed to generate more reliable pseudo-labels based on the similarity of node features. Extensive experiments on six benchmark datasets demonstrate the superiority of MFGCN over state-of-the-art classification methods.(c) 2022 Elsevier Ltd. All rights reserved.
Keyword ：
Pseudo -label supervision Pseudo -label supervision Graph convolutional networks Graph convolutional networks Semi -supervised learning Semi -supervised learning Node classification Node classification
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Yang, Yachao , Sun, Yanfeng , Ju, Fujiao et al. Multi-graph Fusion Graph Convolutional Networks with pseudo-label supervision [J]. | NEURAL NETWORKS , 2023 , 158 : 305-317 . |
| MLA | Yang, Yachao et al. "Multi-graph Fusion Graph Convolutional Networks with pseudo-label supervision" . | NEURAL NETWORKS 158 (2023) : 305-317 . |
| APA | Yang, Yachao , Sun, Yanfeng , Ju, Fujiao , Wang, Shaofan , Gao, Junbin , Yin, Baocai . Multi-graph Fusion Graph Convolutional Networks with pseudo-label supervision . | NEURAL NETWORKS , 2023 , 158 , 305-317 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
Graph Convolutional Network (GCN) is a powerful model for graph representation learning. Since GCN updates nodes with a recursive neighbor aggregation scheme, training GCN on large-scale graphs suffers from enormous computational cost and large memory requirement. The subgraph sampling method trains GCN on sampled small-scale subgraphs to speed up GCN. However, they also suffer from problems, such as training GCN on unconnected and scale-unbalanced subgraphs, which reduce performance and efficiency. Moreover, existing subgraph sampling methods train GCN on subgraphs independently and ignore the relation information among different subgraphs. This paper proposes a novel subgraph sampling method, Improved Adaptive Neighbor Sampling (IANS), and a novel loss function, Subgraph Contrastive Loss. Subgraphs sampled by the IANS method are scale-balanced, inside nodes are significantly relevant, and the sample ratio controls the sparsity of subgraphs. To recover the lost relation information between different subgraphs, the Subgraph Contrastive Loss is defined, which constrains the initially connected nodes in different subgraphs to be closer and pushes unconnected nodes far away in feature space. A series of experiments are conducted, which train GCN with IANS and Subgraph Contrastive Loss for node classification on three different scale datasets. The training time and classification accuracy demonstrate the effectiveness of the proposed method.
Keyword ：
Graph convolutional network Graph convolutional network Subgraph contrastive loss Subgraph contrastive loss Subgraph sampling Subgraph sampling Large-scale graph Large-scale graph Node classification Node classification
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | Zhang, Qi , Sun, Yanfeng , Hu, Yongli et al. A subgraph sampling method for training large-scale graph convolutional network [J]. | INFORMATION SCIENCES , 2023 , 649 . |
| MLA | Zhang, Qi et al. "A subgraph sampling method for training large-scale graph convolutional network" . | INFORMATION SCIENCES 649 (2023) . |
| APA | Zhang, Qi , Sun, Yanfeng , Hu, Yongli , Wang, Shaofan , Yin, Baocai . A subgraph sampling method for training large-scale graph convolutional network . | INFORMATION SCIENCES , 2023 , 649 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
本发明公开了基于多层注意力的视觉定位方法，该方法基于三个模块实现：1)属性注意模块：提取目标对象的细粒度的属性信息；2)上下文注意模块：提取目标对象的周围环境信息；3)匹配模块：结合上两个模块提取到的视觉信息与文本信息匹配找到目标对象。根据文本指导编码与文本语义信息一致的视觉信息来与文本更好的匹配，其包括局部注意力与全局注意力，局部注意力通过跨模态交互提取目标对象细粒度的属性信息；全局注意力通过建立文本为指导的图卷积模型抽取目标对象的上下文信息。两个注意力的结合可以全方位的抽取不同角度的视觉信息，来与文本信息更好的匹配。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 孙艳丰 , 张云茹 , 胡永利 et al. 基于多层注意力的视觉定位方法 : CN202211492369.X[P]. | 2022-11-25 . |
| MLA | 孙艳丰 et al. "基于多层注意力的视觉定位方法" : CN202211492369.X. | 2022-11-25 . |
| APA | 孙艳丰 , 张云茹 , 胡永利 , 姜华杰 , 尹宝才 . 基于多层注意力的视觉定位方法 : CN202211492369.X. | 2022-11-25 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
从原始图像学习的图卷积聚类方法及装置，对预先提供的图的质量要求不高，模型学习的灵活性更强，对各种复杂的聚类应用的鲁棒性更好。方法包括：(1)采用深度自动编码器DAE学习原始图像中判别力强的数据表示，探索数据的潜在属性信息；(2)从原始图中学习具有低秩和稀疏结构的自适应图；引入图的拉普拉斯约束，同时学习最优的图关系和数据的判别表示；(3)进行自监督的聚类，自监督节点表示的学习，在聚类中使用软赋值分布作为聚类标签。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 尹宝才 , 赵珈艺 , 孙艳丰 . 从原始图像学习的图卷积聚类方法及装置 : CN202210543268.4[P]. | 2022-05-16 . |
| MLA | 尹宝才 et al. "从原始图像学习的图卷积聚类方法及装置" : CN202210543268.4. | 2022-05-16 . |
| APA | 尹宝才 , 赵珈艺 , 孙艳丰 . 从原始图像学习的图卷积聚类方法及装置 : CN202210543268.4. | 2022-05-16 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
本发明涉及一种基于对偶动态时空图卷积的交通预测方法，用于解决当前基于图网络的交通预测方法中存在缺少对边建模以及动态建模导致的预测精度不高的问题。首先输入历史交通数据，送到输入层进行处理，然后将输入层的输出送入动态时空层，经过动态时空层中多个堆叠的对偶动态时空块进行时空相关性特征抽取，再将这些特征输入到输出层，最后输出的即是最终的预测结果。其中，最核心和关键的对偶动态时空块包括动态图卷积模块、动态超图卷积模块，以及两个之间的动态交互模块。本发明能很好的挖掘交通数据中复杂的时空相关性，从而揭示动态交通系统潜在的时空关联，进而更加准确的对城市交通数据进行预测。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 孙艳丰 , 江相衡 , 胡永利 et al. 一种基于对偶动态时空图卷积的交通预测方法 : CN202210096933.X[P]. | 2022-01-26 . |
| MLA | 孙艳丰 et al. "一种基于对偶动态时空图卷积的交通预测方法" : CN202210096933.X. | 2022-01-26 . |
| APA | 孙艳丰 , 江相衡 , 胡永利 , 郭侃 , 尹宝才 . 一种基于对偶动态时空图卷积的交通预测方法 : CN202210096933.X. | 2022-01-26 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
一种基于多尺度动态图卷积网络的弱监督人群计数方法属于人群计数在公共安全、城市规划和交通调度等领域。由于交通场景的复杂性和多样性，对大量人群进行点级标注非常困难，而且需要大量人力。弱监督人群计数更适合这些场景，因为它们只需要计数级别的注释。现有的弱监督人群计数忽略了交叉距离人群密度分布的不均匀性和多尺度人群头部，无法获得与全监督人群计数方法相似的准确计数结果。本发明提出了一种多级区域动态图卷积模块来提取不同人群区域之间的内在关系，从而学习动态区域得分，进而优化区域特征表示，还设计了一个粗粒度的多级特征融合模块来提取多尺度人群头部信息。本发明具有较高的回归精度的端到端人群计数能力。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 张勇 , 苗壮壮 , 孙艳丰 et al. 一种基于多尺度动态图卷积的弱监督人群计数方法 : CN202210859858.8[P]. | 2022-07-21 . |
| MLA | 张勇 et al. "一种基于多尺度动态图卷积的弱监督人群计数方法" : CN202210859858.8. | 2022-07-21 . |
| APA | 张勇 , 苗壮壮 , 孙艳丰 , 胡永利 , 尹宝才 . 一种基于多尺度动态图卷积的弱监督人群计数方法 : CN202210859858.8. | 2022-07-21 . |
| Export to | NoteExpress RIS BibTex |
Abstract ：
基于交错空洞卷积UNet的图像语义分割方法适用于计算机视觉领域。该方法使用了交错空洞卷积模块以及边缘保持模块。交错空洞卷积模块通过交错式的融合方式，即避免了将表达不同物体的特征进行相加，又有效的融合分布在不同通道的特征。而边缘保持模块通过将不同卷积核的卷积层的输出做差得到边缘特征，经整合后加和到输出特征上。该模块具有锐化特征边缘信息的能力。从而增强模型对于边缘的预测能力。该方法的提出，主要解决的技术问题包括多感受野的特征融合与图像语义分割的细节优化，从而获得更好的语义分割性能。
Cite：
Copy from the list or Export to your reference management。
| GB/T 7714 | 王少帆 , 刘玉坤 , 孙艳丰 et al. 基于交错空洞卷积UNet的图像语义分割方法 : CN202211106328.2[P]. | 2022-09-11 . |
| MLA | 王少帆 et al. "基于交错空洞卷积UNet的图像语义分割方法" : CN202211106328.2. | 2022-09-11 . |
| APA | 王少帆 , 刘玉坤 , 孙艳丰 , 尹宝才 . 基于交错空洞卷积UNet的图像语义分割方法 : CN202211106328.2. | 2022-09-11 . |
| Export to | NoteExpress RIS BibTex |
Export
| Results： |
Selected to |
| Format： |
