Filters








18,548 Hits in 3.7 sec

Ordered Subgraph Aggregation Networks [article]

Chendi Qian, Gaurav Rattan, Floris Geerts, Christopher Morris, Mathias Niepert
2022 arXiv   pre-print
non-data-driven subgraph-enhanced graph neural networks while reducing computation time.  ...  Numerous subgraph-enhanced graph neural networks (GNNs) have emerged recently, provably boosting the expressive power of standard (message-passing) GNNs.  ...  Essentially, the k-OSWL labels or marks ordered subgraphs and then executes 1-WL on top of the marked graphs, followed by an aggregation phase.  ... 
arXiv:2206.11168v2 fatcat:y7y5glkspzdclopkgw3ikrk4je

Robust android malware detection based on subgraph network and denoising GCN network

Xiaofeng Lu, Jinglun Zhao, Pietro Lio
2022 Proceedings of the 20th Annual International Conference on Mobile Systems, Applications and Services  
The model uses the subgraph network to detect the underlying structural features of the FCG and discover the confusion attack.  ...  A denoising graph neural network is applied to graph convolution to reduce the impact of obfuscation attacks.  ...  For second-order SGNs, open triangles (with at least 2 edges) are defined as the basic subgraph units for building second-order subgraph networks, SGN (2) .  ... 
doi:10.1145/3498361.3538778 fatcat:remvqssjjnajrerporfygjnaey

HoNVis: Visualizing and Exploring Higher-Order Networks [article]

Jun Tao, Jian Xu, Chaoli Wang, Nitesh V. Chawla
2017 arXiv   pre-print
Unlike the conventional first-order network (FoN), the higher-order network (HoN) provides a more accurate description of transitions by creating additional nodes to encode higher-order dependencies.  ...  In this paper, we present HoNVis, a novel visual analytics framework for exploring higher-order dependencies of the global ocean shipping network.  ...  subgraph and the entire network.  ... 
arXiv:1702.00737v1 fatcat:2rl3yxp3tzhele6kwfw3sjin6u

MSGCN: Multi-Subgraph Based Heterogeneous Graph Convolution Network Embedding

Junhui Chen, Feihu Huang, Jian Peng
2021 Applied Sciences  
Finally, the node representations are obtained by aggregating the representation vectors of nodes in each subgraph.  ...  So we proposed a new method called Multi-Subgraph based Graph Convolution Network (MSGCN), which uses topology information, semantic information, and node feature information to learn node embedding vector  ...  In order to address the above problems, this paper proposes a new model to deal with heterogeneous graphs, called MSGCN (Multi-Subgraph based Graph Convolution Network).  ... 
doi:10.3390/app11219832 fatcat:jllatfdxlzdvjd5sqeejuy6twe

Parallel discovery of network motifs

Pedro Ribeiro, Fernando Silva, Luís Lopes
2012 Journal of Parallel and Distributed Computing  
Many natural structures can be naturally represented by complex networks.  ...  Our strategies are capable of dealing both with exact and approximate network motif discovery.  ...  This is a huge a amount of data that we need to aggregate in order to calculate the subgraph significance. Note that the number of possible k-subgraphs grows super-exponentially as k increases.  ... 
doi:10.1016/j.jpdc.2011.08.007 fatcat:cot76i6rpbaejb2yqnl2243vem

The topological relationship between the large-scale attributes and local interaction patterns of complex networks

A. Vazquez, R. Dobrin, D. Sergi, J.- P. Eckmann, Z. N. Oltvai, A.- L. Barabasi
2004 Proceedings of the National Academy of Sciences of the United States of America  
but must naturally aggregate into subgraph clusters.  ...  networks.  ...  but must naturally aggregate into subgraph clusters.  ... 
doi:10.1073/pnas.0406024101 pmid:15598746 pmcid:PMC539752 fatcat:xmxnkzwkjreszdlmpo4dbka4fy

Atomic subgraphs and the statistical mechanics of networks [article]

Anatol E. Wegner, Sofia Olhede
2020 arXiv   pre-print
These models include random graphs with arbitrary distributions of subgraphs, random hypergraphs, bipartite models, stochastic block models, models of multilayer networks and their degree corrected and  ...  This allows the for the generation of graphs with extensive numbers of triangles and other network motifs commonly observed in many real world networks.  ...  Higher order atoms and network communities Including higher order structures such as triangles in generative models can have a significant impact on inference of network communities.  ... 
arXiv:2008.10346v1 fatcat:ti6g26ouxzawxmd5ojcbutiaxe

Neural Subgraph Matching [article]

Rex Ying, Zhaoyu Lou, Jiaxuan You, Chengtao Wen, Arquimedes Canedo, Jure Leskovec
2020 arXiv   pre-print
NeuroMatch decomposes query and target graphs into small subgraphs and embeds them using graph neural networks.  ...  Despite being an NP-complete problem, the subgraph matching problem is crucial in domains ranging from network science and database systems to biochemistry and cognitive science.  ...  NeuroMatch trains a graph neural network to learn the order embedding, and uses a max-margin loss to ensure that the subgraph relationships are captured.  ... 
arXiv:2007.03092v2 fatcat:iukf3larbbdt5mgyblw3lca5cq

Maximal prime subgraph decomposition of Bayesian networks

K.G. Olesen, A.L. Madsen
2002 IEEE Transactions on Systems Man and Cybernetics Part B (Cybernetics)  
In this paper we present a method for decomposition of Bayesian networks into their maximal prime subgraphs.  ...  We also identify a number of tasks performed on Bayesian networks that can benefit from maximal prime subgraph decomposition.  ...  Further research is needed in order to draw precise conclusions on the importance of exploiting maximal prime subgraph decompositions.  ... 
doi:10.1109/3477.979956 pmid:18238100 fatcat:vg4qovtxszbshcnu2lwrq33ikq

Robust Hierarchical Graph Classification with Subgraph Attention [article]

Sambaran Bandyopadhyay, Manasvi Aggarwal, M. Narasimha Murty
2020 arXiv   pre-print
Attention mechanism applied on the neighborhood of a node improves the performance of graph neural networks.  ...  Graph neural networks get significant attention for graph representation and classification in machine learning community.  ...  Besides, recent literature also shows that higher order GNNs that directly aggregate features from higher order neighborhood of a node are theoretically more powerful than 1st order GNNs [24] .  ... 
arXiv:2007.10908v1 fatcat:h6jbr6ke5jclhpofwvy64k6sey

VCExplorer: A Interactive Graph Exploration Framework Based on Hub Vertices with Graph Consolidation [article]

Huiju Wang, Zhengkui Wang, Kian-Lee Tan, Chee-Yong Chan, Qi Fan, Xiao Yue
2017 arXiv   pre-print
In addition, we propose efficient graph aggregation algorithms over multiple subgraphs via computation sharing.  ...  Graphs have been widely used to model different information networks, such as the Web, biological networks and social networks (e.g. Twitter).  ...  INTRODUCTION Graphs are powerful tools to model a variety of information networks, such as the Web, biological networks and social networks (e.g. Twitter).  ... 
arXiv:1709.06745v1 fatcat:z6nvkc7xnffipf4cax3dutydqm

A Multi-Scale Approach for Graph Link Prediction

Lei Cai, Shuiwang Ji
2020 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In this work, we propose a novel node aggregation method that can transform the enclosing subgraph into different scales and preserve the relationship between two target nodes for link prediction.  ...  Graphs in different scales can provide scale-invariant information, which enables graph neural networks to learn invariant features and improve link prediction performance.  ...  By aggregating the subgraph iteratively, we can obtain a series of subgraphs in different scales.  ... 
doi:10.1609/aaai.v34i04.5731 fatcat:7zgj6kjmxzahvjz724k2fu2u7m

A Self-adaptive Fault-Tolerant Mechanism in Wireless Sensor Networks [chapter]

Wei Xiao, Ming Xu, Yingwen Chen
2009 Lecture Notes of the Institute for Computer Sciences, Social Informatics and Telecommunications Engineering  
Moreover, we discussed and tried to solve the problems on FTMAWSS (fault-tolerant maximum average-weighted spanning subgraph) in weighted connected graph initiating from clustering WSNs.  ...  It is crucial for the sensors to form an optimal network topology and tune to transmission attempt rates in a way that optimize network throughput.  ...  So the self-adaptive of sensor networks can be used to construct an optimal network subgraph with fault tolerant in this paper in order to form an effective, practical subgraph topology.  ... 
doi:10.1007/978-3-642-10485-5_17 fatcat:hcgexo7uqfdmrm3boafsdwyrmy

Soft-mask: Adaptive Substructure Extractions for Graph Neural Networks [article]

Mingqi Yang, Yanming Shen, Heng Qi, Baocai Yin
2022 arXiv   pre-print
A graph neural network should be able to efficiently extract task-relevant structures and be invariant to irrelevant parts, which is challenging for general message passing GNNs.  ...  more flexible to extract subgraphs with arbitrary sizes.  ...  The soft-mask GNN layer applies continuous mask values in order to maintain the differentiability of the networks.  ... 
arXiv:2206.05499v1 fatcat:kzv2bi3qwnby7a4pwscpqtsysy

Metapath- and Entity-aware Graph Neural Network for Recommendation [article]

Muhammad Umer Anwaar, Zhiwei Han, Shyam Arumugaswamy, Rayyan Ahmad Khan, Thomas Weber, Tianming Qiu, Hao Shen, Yuanting Liu, Martin Kleinsteuber
2021 arXiv   pre-print
We propose metaPath and Entity-Aware Graph Neural Network (PEAGNN), which trains multilayer GNNs to perform metapath-aware information aggregation on such subgraphs.  ...  In graph neural networks (GNNs), message passing iteratively aggregates nodes' information from their direct neighbors while neglecting the sequential nature of multi-hop node connections.  ...  NFM [14] utilizes neural networks to enhance high-order feature interactions with non-linearity. As suggested by He et al. [14] , we apply one hidden layer neural network on input features. 2.  ... 
arXiv:2010.11793v3 fatcat:xn6j6yhyevdcdmssxpgmo4cwcq
« Previous Showing results 1 — 15 out of 18,548 results