9,159 Hits in 3.6 sec

Search to aggregate neighborhood for graph neural network [article]

Huan Zhao, Quanming Yao, Weiwei Tu
2021 arXiv   pre-print
Recent years have witnessed the popularity and success of graph neural networks (GNN) in various scenarios.  ...  In this work, to obtain the data-specific GNN architectures and address the computational challenges facing by NAS approaches, we propose a framework, which tries to Search to Aggregate NEighborhood (SANE  ...  Terms-graph neural network, neural architecture search, message passing I.  ... 
arXiv:2104.06608v2 fatcat:xsjiezgfajhihhld5munmkbvqa

Evolutionary Architecture Search for Graph Neural Networks [article]

Min Shi, David A.Wilson, Xingquan Zhu, Yu Huang, Yuan Zhuang, Jianxun Liu, Yufei Tang
2020 arXiv   pre-print
However, very litter work has been done about Graph Neural Networks (GNN) learning on unstructured network data.  ...  To the best of our knowledge, this is the first work to introduce and evaluate evolutionary architecture search for GNN models.  ...  CONCLUSION In this paper, we aim to demonstrate the effectiveness of evolutionary neural architecture search for optimizing graph neural network models on graph-structured data.  ... 
arXiv:2009.10199v1 fatcat:k2h23byz2jfebbiw3mzckveblm

Simplifying Architecture Search for Graph Neural Network [article]

Huan Zhao and Lanning Wei and Quanming Yao
2020 arXiv   pre-print
To overcome these drawbacks, we propose the SNAG framework (Simplified Neural Architecture search for Graph neural networks), consisting of a novel search space and a reinforcement learning based search  ...  Recent years have witnessed the popularity of Graph Neural Networks (GNN) in various scenarios.  ...  neural networks (CNN) and recurrent neural networks (RNN).  ... 
arXiv:2008.11652v2 fatcat:7mgbknvplvcsxoh3mcs3ofhduy

RAW-GNN: RAndom Walk Aggregation based Graph Neural Network [article]

Di Jin, Rui Wang, Meng Ge, Dongxiao He, Xiang Li, Wei Lin, Weixiong Zhang
2022 arXiv   pre-print
It replaces the conventional neighborhoods with path-based neighborhoods and introduces a new path-based aggregator based on Recurrent Neural Networks.  ...  Here, we introduce a novel aggregation mechanism and develop a RAndom Walk Aggregation-based Graph Neural Network (called RAW-GNN) method.  ...  We integrate random walk sampling into graph neural networks and extend the conventional neighborhoods to k-hop path-based neighborhoods.  ... 
arXiv:2206.13953v1 fatcat:5ok76d4zhvb2zhrpft4olunoxi

Policy-GNN: Aggregation Optimization for Graph Neural Networks [article]

Kwei-Herng Lai, Daochen Zha, Kaixiong Zhou, Xia Hu
2020 arXiv   pre-print
Recently, increasing attention has been paid on graph neural networks (GNNs), which aim to model the local graph structures and capture the hierarchical patterns by aggregating the information from neighbors  ...  It is a challenging task to develop an effective aggregation strategy for each node, given complex graphs and sparse features.  ...  Learning Graph Representations with Deep Neural Networks Due to the great success of convolution neural networks on image data, graph neural networks [5, 14, 20, 37] have been extensively studied for  ... 
arXiv:2006.15097v1 fatcat:b5h7cvwp5rh2xcinb7p4rg4cia

Geom-GCN: Geometric Graph Convolutional Networks [article]

Hongbin Pei, Bingzhe Wei, Kevin Chen-Chuan Chang, Yu Lei, Bo Yang
2020 arXiv   pre-print
From the observations on classical neural network and network geometry, we propose a novel geometric aggregation scheme for graph neural networks to overcome the two weaknesses.  ...  Message-passing neural networks (MPNNs) have been successfully applied to representation learning on graphs in a variety of real-world applications.  ...  ACKNOWLEDGMENTS We thank the reviewers for their valuable feedback. This work was supported in part by National Natural Science  ... 
arXiv:2002.05287v2 fatcat:yf7xylymlzbgjadk726bvr3pfu

Rethinking Graph Neural Architecture Search from Message-passing [article]

Shaofei Cai, Liang Li, Jincan Deng, Beichen Zhang, Zheng-Jun Zha, Li Su, Qingming Huang
2021 arXiv   pre-print
Graph neural networks (GNNs) emerged recently as a standard toolkit for learning from data on graphs.  ...  message-passing mechanism to construct powerful graph network search space.  ...  Figure 1 shows an example of a graph neural network searched by GNAS.  ... 
arXiv:2103.14282v4 fatcat:rshdr2tqvbbm7ijborgvtjju7a

Graph Representation Learning for Road Type Classification

Zahra Gharaee, Shreyas Kowshik, Oliver Stromann, Michael Felsberg
2021 Pattern Recognition  
We present a novel learning-based approach to graph representations of road networks employing state-of-the-art graph convolutional neural networks.  ...  Furthermore, we propose a novel aggregation approach, Graph Attention Isomorphism Network, GAIN.  ...  Using generative models based on deep neural networks have also been extensively studied for graph generations.  ... 
doi:10.1016/j.patcog.2021.108174 fatcat:funo4akd45f2xmf3ro6m6vl4xu

Enhance Information Propagation for Graph Neural Network by Heterogeneous Aggregations [article]

Dawei Leng, Jinjiang Guo, Lurong Pan, Jie Li, Xinyu Wang
2021 arXiv   pre-print
Tens of different graph neural network variants have been proposed, most following a neighborhood aggregation scheme, where the node features are updated via aggregating features of its neighboring nodes  ...  Graph neural networks are emerging as continuation of deep learning success w.r.t. graph data.  ...  Model Structure With the generic neighborhood aggregation layer as eq.7 and READOUT layer as eq.8, we build our graph neural network HAG-Net for graph-level tasks by stacking multiple neighborhood aggregation  ... 
arXiv:2102.04064v1 fatcat:o6nchngbkvdyvmm5riksdmlmzi

Learning Better Representations for Neural Information Retrieval with Graph Information

Xiangsheng Li, Maarten de Rijke, Yiqun Liu, Jiaxin Mao, Weizhi Ma, Min Zhang, Shaoping Ma
2020 Proceedings of the 29th ACM International Conference on Information & Knowledge Management  
In this study, we aim to incorporate this rich information encoded in these two graphs into existing neural ranking models.  ...  We present two graph-based neural ranking models (EmbRanker and AggRanker) to enrich learned text representations with graph information that captures rich users' interaction behavior information.  ...  propagating the information of all the neighborhoods to the focal node by a graph neural network (GNN).  ... 
doi:10.1145/3340531.3411957 dblp:conf/cikm/LiR0MM0M20 fatcat:vurrhgyzpzau5epajhzyz5524a

ATPGNN: Reconstruction of Neighborhood in Graph Neural Networks with Attention-based Topological Patterns

Kehao Wang, Hantao Qian, Xuming Zeng, Mozi Chen, Kezhong Liu, Kai Zheng, Pan Zhou, Dapeng Wu
2021 IEEE Access  
Graph Neural Networks (GNNs) have been applied in many fields of semi-supervised node classification for non-Euclidean data.  ...  Third, we use some network embedding methods to get graph structure information of each node.  ...  For the aggregation method of topology neighborhood, we adopt the basic settings of the graph neural network method used in our model.  ... 
doi:10.1109/access.2021.3050541 fatcat:pn5uyqnvr5dw3ord6hhytzlgqm

GraphNAS: Graph Neural Architecture Search with Reinforcement Learning [article]

Yang Gao, Hong Yang, Peng Zhang, Chuan Zhou, Yue Hu
2019 arXiv   pre-print
In this paper, we propose a Graph Neural Architecture Search method (GraphNAS for short) that enables automatic search of the best graph neural architecture based on reinforcement learning.  ...  Graph Neural Networks (GNNs) have been popularly used for analyzing non-Euclidean data such as social network data and biological data.  ...  Introduction Graph Neural Networks (GNNs) have been popularly used for analyzing graph data such as social network data and biological data.  ... 
arXiv:1904.09981v2 fatcat:5pyltylpyrf7rhaqhrm32xlmuy

A review of recommendation system research based on bipartite graph

Ziteng Wu, Chengyun Song, Yunqing Chen, Lingxuan Li, I. Barukčić
2021 MATEC Web of Conferences  
An review of the recommendation system of graphs summarizes the three characteristics of graph neural network processing bipartite graph data in the recommendation field: interchangeability, Multi-hop  ...  The biggest contribution of the full paper is that it summarizes the general framework of graph neural network processing bipartite graph recommendation from the models with the best recommendation effect  ...  The role of this layer is to learn a lowdimensional vector representation for the input of the graph neural network model.  ... 
doi:10.1051/matecconf/202133605010 fatcat:qyiiqjnev5eublqlmps4gcworq

GAIN: Graph Attention Interaction Network for Inductive Semi-Supervised Learning over Large-scale Graphs [article]

Yunpeng Weng and Xu Chen and Liang Chen and Wei Liu
2020 arXiv   pre-print
In this paper, we propose a novel graph neural network architecture, Graph Attention \& Interaction Network (GAIN), for inductive learning on graphs.  ...  Graph neural network models generate node embeddings by merging nodes features with the aggregated neighboring nodes information.  ...  To address this challenging problem, we propose a novel graph neural network model, Graph Attention & Interaction Network (GAIN).  ... 
arXiv:2011.01393v1 fatcat:ahcjc4bm5nhulhg3k5vpg5r3jm

Tree Decomposed Graph Neural Network [article]

Yu Wang, Tyler Derr
2021 arXiv   pre-print
deeper graph neural networks.  ...  Graph Neural Networks (GNNs) have achieved significant success in learning better representations by performing feature propagation and transformation iteratively to leverage neighborhood information.  ...  Graph Neural Networks Typically, most graph neural networks (GNNs) can be decomposed into two operational procedures: (1) neighborhood propagation and aggregation, and (2) feature transformation.  ... 
arXiv:2108.11022v1 fatcat:5vnekq7t7zh2bkdrulg2zfgq6m
« Previous Showing results 1 — 15 out of 9,159 results