332 Hits in 4.5 sec

CoSimGNN: Towards Large-scale Graph Similarity Computation [article]

Haoyan Xu, Runjian Chen, Yunsheng Bai, Ziheng Duan, Jie Feng, Yizhou Sun, Wei Wang
2021 arXiv   pre-print
In this paper, we focus on similarity computation for large-scale graphs and propose the "embedding-coarsening-matching" framework, which first embeds and coarsens large graphs to coarsened graphs with  ...  denser local topology and then deploys fine-grained interactions on the coarsened graphs for the final similarity scores.  ...  Similarity Computation via Graph Neural Networks.  ... 
arXiv:2005.07115v6 fatcat:azntqlddurg5blf47qzttiqsiq

Pyramidal Graph Echo State Networks

Filippo Maria Bianchi, Claudio Gallicchio, Alessio Micheli
2020 The European Symposium on Artificial Neural Networks  
We analyze graph neural network models that combine iterative message-passing implemented by a function with untrained weights and graph pooling operations.  ...  In particular, we alternate randomized neural message passing with graph coarsening operations, which provide multiple views of the underlying graph.  ...  The potentiality of hierarchical GESN architectures in designing fast and deep models for graph classification has been recently shown in [6] .  ... 
dblp:conf/esann/BianchiGM20 fatcat:u33sfufrpreh7ex7icgsvoau7a

Multivariate Time Series Classification with Hierarchical Variational Graph Pooling [article]

Ziheng Duan, Haoyan Xu, Yueyang Wang, Yida Huang, Anni Ren, Zhongbin Xu, Yizhou Sun, Wei Wang
2021 arXiv   pre-print
Then we combine GNNs and our proposed variational graph pooling layers for joint graph representation learning and graph coarsening, after which the graph is progressively coarsened to one node.  ...  We first convert MTS slices to graphs by utilizing interactions of variables via graph structure learning module and attain the spatial-temporal graph node features via temporal convolutional module.  ...  MTPool is an end-to-end joint framework for graph structure learning, temporal convolution, graph representation learning and graph coarsening.  ... 
arXiv:2010.05649v2 fatcat:24qq5zhua5a2rhfc5wf6mp6wza

Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering [article]

Michaël Defferrard, Xavier Bresson, Pierre Vandergheynst
2017 arXiv   pre-print
Experiments on MNIST and 20NEWS demonstrate the ability of this novel deep learning system to learn local, stationary, and compositional features on graphs.  ...  Importantly, the proposed technique offers the same linear computational complexity and constant learning complexity as classical CNNs, while being universal to any graph structure.  ...  accuracies of the proposed graph CNN and a classical CNN on MNIST. tion on mesh patches, and formulated a deep learning architecture which allows comparison across different manifolds.  ... 
arXiv:1606.09375v3 fatcat:n3dutxnbrrhrxnc5ywzwamah6u

Graph Neural Networks: Taxonomy, Advances and Trends [article]

Yu Zhou, Haixia Zheng, Xin Huang, Shufeng Hao, Dengao Li, Jumin Zhao
2022 arXiv   pre-print
It is expected that more and more scholars can understand and exploit the graph neural networks, and use them in their research community.  ...  This survey aims to overcome this limitation, and provide a comprehensive review on the graph neural networks.  ...  The literature [42] proposes a novel deep Hierarchical Graph Convolutional Network (H-GCN) consisting of graph coarsening layers and graph refining layers.  ... 
arXiv:2012.08752v3 fatcat:xj2kambrabfj3g5ldenfyixzu4

Graph Embedding via Graph Summarization

Jingyanning Yang, Jinguo You, Xiaorong Wan
2021 IEEE Access  
Graph representation learning aims to represent the structural and semantic information of graph objects as dense real value vectors in low dimensional space by machine learning.  ...  However, directly computing the embeddings for original graphs is prohibitively inefficient, especially for large-scale graphs.  ...  [6] proposed a fast graph embedding via a coarsening algorithm based on Schur complements for computing the vertices' embeddings.  ... 
doi:10.1109/access.2021.3067901 fatcat:6h2xmr42cbgztmazzj7o6xzodu

Hierarchical Representation Learning in Graph Neural Networks with Node Decimation Pooling [article]

Filippo Maria Bianchi, Daniele Grattarola, Lorenzo Livi, Cesare Alippi
2020 arXiv   pre-print
In graph neural networks (GNNs), pooling operators compute local summaries of input graphs to capture their global properties, and they are fundamental for building deep GNNs that learn hierarchical representations  ...  During training, the GNN learns new node representations and fits them to a pyramid of coarsened graphs, which is computed offline in a pre-processing stage. NDP consists of three steps.  ...  INTRODUCTION Generating hierarchical representations across the layers of a neural network is key to deep learning methods.  ... 
arXiv:1910.11436v2 fatcat:cxawupuzbzcylj7z7ilfrtaewi

Graph-level Neural Networks: Current Progress and Future Directions [article]

Ge Zhang, Jia Wu, Jian Yang, Shan Xue, Wenbin Hu, Chuan Zhou, Hao Peng, Quan Z. Sheng, Charu Aggarwal
2022 arXiv   pre-print
However, with the increasing scale and complexity of graphs, Graph-level Neural Networks (GLNNs, deep learning-based graph-level learning methods) have been attractive due to their superiority in modeling  ...  To frame this survey, we propose a systematic taxonomy covering GLNNs upon deep neural networks, graph neural networks, and graph pooling.  ...  Hierarchical Pooling Hierarchical pooling learns coarsen-grained graph structures by gradually applying the pooling layer to downsize the graph.  ... 
arXiv:2205.15555v1 fatcat:ecjtm2ezrndu7jr5ijninkab44

Graph Coarsening with Neural Networks [article]

Chen Cai, Dingkang Wang, Yusu Wang
2021 arXiv   pre-print
In this work, we leverage the recent progress of deep learning on graphs for graph coarsening.  ...  As large-scale graphs become increasingly more prevalent, it poses significant computational challenges to process, extract and analyze large graph data.  ...  Unsupervised learning of graph hierarchical abstractions with differen- tiable coarsening and optimal transport. arXiv preprint arXiv:1912.11176, 2019. Tengfei Ma, Jie Chen, and Cao Xiao.  ... 
arXiv:2102.01350v1 fatcat:mslf2m5gu5dr5k3c7dz6nx7i5i

Shortest Path Distance Approximation Using Deep Learning Techniques

Fatemeh Salehi Rizi, Joerg Schloetterer, Michael Granitzer
2018 2018 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining (ASONAM)  
Computing shortest path distances between nodes lies at the heart of many graph algorithms and applications.  ...  In this paper, we utilize vector embeddings learnt by deep learning techniques to approximate the shortest paths distances in large graphs.  ...  While HARP coarsens the graph and learns representations via hierarchically collapsed graphs, Walklets skips over steps in the random walks.  ... 
doi:10.1109/asonam.2018.8508763 dblp:conf/asunam/RiziSG18 fatcat:7oxiop5lgrbu7oytwouraxdksy

Survey on graph embeddings and their applications to machine learning problems on graphs

Ilya Makarov, Dmitrii Kiselev, Nikita Nikitinsky, Lovro Subelj
2021 PeerJ Computer Science  
First, we start with the methodological approach and extract three types of graph embedding models based on matrix factorization, random-walks and deep learning approaches.  ...  compression, and a family of the whole graph embedding algorithms suitable for graph classification, similarity and alignment problems.  ...  ., 2018) with graph partitioning and coarsening to provide fast embedding computation on the GPU.  ... 
doi:10.7717/peerj-cs.357 pmid:33817007 pmcid:PMC7959646 fatcat:ntronyrbgfbedez5dks6h4hoq4

Temporal Multiresolution Graph Neural Networks For Epidemic Prediction [article]

Truong Son Hy and Viet Bach Nguyen and Long Tran-Thanh and Risi Kondor
2022 arXiv   pre-print
In this paper, we introduce Temporal Multiresolution Graph Neural Networks (TMGNN), the first architecture that both learns to construct the multiscale and multiresolution graph structures and incorporates  ...  epidemic in several European countries, and have obtained competitive results in comparison to other previous state-of-the-art temporal architectures and graph learning algorithms.  ...  In the case of graph property regression, we want MGN to learn to predict a real value y ∈ R for each graph G while learning to construct a hierarchical structure of latents and coarsen graphs.  ... 
arXiv:2205.14831v3 fatcat:5iol6griyzadxpuuaw2j73doo4

MathNet: Haar-Like Wavelet Multiresolution-Analysis for Graph Representation and Learning [article]

Xuebin Zheng, Bingxin Zhou, Ming Li, Yu Guang Wang, Junbin Gao
2021 arXiv   pre-print
The whole workflow could be formed with a multi-level graph analysis, which not only helps embed the intrinsic topological information of each graph into the GNN, but also supports fast computation of  ...  The hierarchical graph pooling layers are then involved to downsample graph resolution while simultaneously remove redundancy within graph signals.  ...  The compressive Haar transform follows a similar fast computational strategy as the adjoint transform in the convolutional layer.  ... 
arXiv:2007.11202v2 fatcat:yxayaaegobf5ddtd67sraisyza

Classifying Signals on Irregular Domains via Convolutional Cluster Pooling [article]

Angelo Porrello, Davide Abati, Simone Calderara, Rita Cucchiara
2019 arXiv   pre-print
We present a novel and hierarchical approach for supervised classification of signals spanning over a fixed graph, reflecting shared properties of the dataset.  ...  Extensive experiments on NTU RGB+D, CIFAR-10 and 20NEWS demonstrate the effectiveness of the proposed technique in capturing both local and global patterns in graph-structured data out of different domains  ...  Acknowledgments The authors thank Ottavia Credi for the assistance she provided with the editing and revision of the paper.  ... 
arXiv:1902.04850v1 fatcat:gu352iadu5ezfalogkwybd6yza

Graph Partitioning and Graph Neural Network based Hierarchical Graph Matching for Graph Similarity Computation [article]

Haoyan Xu, Ziheng Duan, Jie Feng, Runjian Chen, Qianru Zhang, Zhongbin Xu, Yueyang Wang
2021 arXiv   pre-print
Graph similarity computation aims to predict a similarity score between one pair of graphs to facilitate downstream applications, such as finding the most similar chemical compounds similar to a query  ...  Recently, some graph similarity computation models based on neural networks have been proposed, which are either based on graph-level interaction or node-level comparison.  ...  Recently, researchers proposed some representative graph deep learning models for graph similarity computation.  ... 
arXiv:2005.08008v3 fatcat:j2otiboex5bzbapvozca5s55xy
« Previous Showing results 1 — 15 out of 332 results