Filters








1,221 Hits in 2.6 sec

Scalable Graph Neural Networks via Bidirectional Propagation [article]

Ming Chen, Zhewei Wei, Bolin Ding, Yaliang Li, Ye Yuan, Xiaoyong Du, Ji-Rong Wen
2021 arXiv   pre-print
Graph Neural Networks (GNN) is an emerging field for learning on non-Euclidean data. Recently, there has been increased interest in designing GNN that scales to large graphs.  ...  This paper presents GBP, a scalable GNN that utilizes a localized bidirectional propagation process from both the feature vectors and the training/testing nodes.  ...  Conclusion This paper presents GBP, a scalable GNN based on localized bidirectional propagation.  ... 
arXiv:2010.15421v3 fatcat:bfyal2ha5vh5rot5z2s5q52t4m

A Robust and Scalable Graph Neural Network for Accurate Single Cell Classification [article]

Yuansong Zeng, Xiang Zhou, Zixiang Pan, Yutong Lu, Yuedong Yang
2021 bioRxiv   pre-print
One powerful way for the transferring is to learn cell relations through the graph neural network (GNN), while vanilla GNN is difficult to process millions of cells due to the expensive costs of the message-passing  ...  To overcome the slow information propagation of GNN at each training epoch, the diffused information is pre-calculated via the approximate Generalized PageRank algorithm, enabling sublinear complexity  ...  Here, we present a scalable graph neural network learning model for cell annotations by constructing the graph via BBKNN, and pre-calculate the diffused features via the graph bidirectional propagation  ... 
doi:10.1101/2021.06.24.449752 fatcat:y47otzsq35gupcya64lrglk4lu

SCALE-Net: Scalable Vehicle Trajectory Prediction Network under Random Number of Interacting Vehicles via Edge-enhanced Graph Convolutional Neural Network [article]

Hyeongseok Jeon, Junwon Choi, Dongsuk Kum
2020 arXiv   pre-print
The SCALE-Net employs the Edge-enhance Graph Convolutional Neural Network (EGCN) for the inter-vehicular interaction embedding network.  ...  Since the proposed EGCN is inherently scalable with respect to the graph node (an agent in this study), the model can be operated independently from the total number of vehicles considered.  ...  In SCALE-Net, EGCN-LSTM based sequential interaction SCALE-Net: Scalable Vehicle Trajectory Prediction Network under Random Number of Interacting Vehicles via Edge-enhanced Graph Convolutional Neural  ... 
arXiv:2002.12609v1 fatcat:hinyd4cjjnht3ekl4mby3tsiem

IEEE Access Special Section Editorial: Advanced Data Mining Methods for Social Computing

Yongqiang Zhao, Shirui Pan, Jia Wu, Huaiyu Wan, Huizhi Liang, Haishuai Wang, Huawei Shen
2020 IEEE Access  
The scalable part is a neural network that can jointly encode, compress, and fuse various types of contexts.  ...  The article by Jiang et al., ''Factorization meets neural networks: A scalable and efficient recommender for solving the new user problem,'' proposes a scalable and efficient recommender to solve the new  ... 
doi:10.1109/access.2020.3043060 fatcat:qbqk5f4ojvadlazhk2mc343sra

GNN-XML: Graph Neural Networks for Extreme Multi-label Text Classification [article]

Daoming Zong, Shiliang Sun
2020 arXiv   pre-print
To overcome these problems, we propose GNN-XML, a scalable graph neural network framework tailored for XMTC problems.  ...  Specifically, we exploit label correlations via mining their co-occurrence patterns and build a label graph based on the correlation matrix.  ...  In this paper, we propose GNN-XML, a scalable graph neural network framework tailored for the XMTC problem.  ... 
arXiv:2012.05860v1 fatcat:l57bt2tkazaf5cdnokpug4e77m

Series Editorial: The Second Issue of the Series on Machine Learning in Communications and Networks

Geoffrey Y. Li, Walid Saad, Ayfer Ozgur, Peter Kairouz, Zhijin Qin, Jakob Hoydis, Zhu Han, Deniz Gunduz, Jaafar Elmirghani
2021 IEEE Journal on Selected Areas in Communications  
The paper "Multi-kernel Clustering via Nonnegative Matrix Factorization Tailored Graph Tensor over Distributed Networks" by Mukherjee et al. seeks to address the problems of multiple kernel graph-based  ...  The paper, titled "Pruning and Quantizing Neural Belief Propagation Decoders," by Buchberger et al., introduces the idea of learning through neural belief propagation a different parity check matrix for  ... 
doi:10.1109/jsac.2021.3078790 fatcat:mg74t6j4bvcd5glxtnzjf2ch44

Star Graph Neural Networks for Session-based Recommendation

Zhiqiang Pan, Fei Cai, Wanyu Chen, Honghui Chen, Maarten de Rijke
2020 Proceedings of the 29th ACM International Conference on Information & Knowledge Management  
We propose Star Graph Neural Networks with Highway Networks (SGNN-HN) for session-based recommendation.  ...  Thus graph neural network (GNN) based models have been proposed to capture the transition relationship between items.  ...  Learning item embeddings on graphs Next, we present how star graph neural networks (SGNN) can propagate information between the nodes of the graph.  ... 
doi:10.1145/3340531.3412014 dblp:conf/cikm/PanCCCR20 fatcat:gi5jjtxocjhl7l5rzmyibpro34

Deep Node Ranking for Neuro-symbolic Structural Node Embedding and Classification [article]

Blaž Škrlj, Jan Kralj, Janez Konc, Marko Robnik-Šikonja, Nada Lavrač
2021 arXiv   pre-print
This paper contributes a novel approach to learning network node embeddings and direct node classification using a node ranking scheme coupled with an autoencoder-based neural network architecture.  ...  The scaling laws associated with DNR were also investigated on 1488 synthetic Erdős-Rényi networks, demonstrating its scalability to tens of millions of links.  ...  For example, ranking was used to prioritize propagation [35] and to scale graph neural networks [36] .  ... 
arXiv:1902.03964v6 fatcat:zlwkh66cqrclpiydbsr2ckdcf4

AB-Net: A Novel Deep Learning Assisted Framework for Renewable Energy Generation Forecasting

Noman Khan, Fath U Min Ullah, Ijaz Ul Haq, Samee Ullah Khan, Mi Young Lee, Sung Wook Baik
2021 Mathematics  
The second step performs deep preprocessing of the acquired data via several de-noising and cleansing filters to clean the data and normalize them prior to actual processing.  ...  To handle this issue, we propose a novel architecture called 'AB-Net': a one-step forecast of RE generation for short-term horizons by incorporating an autoencoder (AE) with bidirectional long short-term  ...  Nomenclature ANN Artificial neural network AE Autoencoder AI Artificial intelligence BiLSTM Bidirectional long short-term memory CNQR Copula-based nonlinear quantile regression CNN Convolutional neural  ... 
doi:10.3390/math9192456 fatcat:7b52vdsmzjd7diyiv6temsxfky

Rethinking Dialogue State Tracking with Reasoning [article]

Lizi Liao, Yunshan Ma, Wenqiang Lei, Tat-Seng Chua
2020 arXiv   pre-print
Toward scalable neural dialogue state tracking model. arXiv preprint arXiv:1812.00899.  ...  After the mapping of beliefs to the database bipartite graph via g(·), we start to do belief propagation over the graph.  ... 
arXiv:2005.13129v2 fatcat:6zemr7ileve4haojiciag4z2vm

Question Answering by Reasoning Across Documents with Graph Convolutional Networks [article]

Nicola De Cao, Wilker Aziz, Ivan Titov
2019 arXiv   pre-print
Graph convolutional networks (GCNs) are applied to these graphs and trained to perform multi-step reasoning.  ...  We introduce a neural model which integrates and reasons relying on information spread within documents and across multiple documents. We frame it as an inference problem on a graph.  ...  This project is supported by SAP Innovation Center Network, ERC Starting Grant BroadSem (678254) and the Dutch Organization for Scientific Research (NWO) VIDI 639.022.518.  ... 
arXiv:1808.09920v3 fatcat:pdd4mc5tkjgezeduqouky2yrba

Deep Feedback Network for Recommendation

Ruobing Xie, Cheng Ling, Yalong Wang, Rui Wang, Feng Xia, Leyu Lin
2020 Proceedings of the Twenty-Ninth International Joint Conference on Artificial Intelligence  
Specifically, we propose a novel Deep feedback network (DFN) modeling click, unclick and dislike behaviors.  ...  The first part is a Communication Agent Network (CAN) with a Message Propagation Graph Neural Network (MPGNN) based active communication network.  ...  Similarly, MaCAR uses Message Propagation based neural networks to model dynamic graphs from traffic. Our Method We first introduce some notations and terms.  ... 
doi:10.24963/ijcai.2020/345 dblp:conf/ijcai/YuLWJHC0020 fatcat:cp5f7rfrrnesxg3o4ttvesllnq

Question Answering by Reasoning Across Documents with Graph Convolutional Networks

Nicola De Cao, Wilker Aziz, Ivan Titov
2019 Proceedings of the 2019 Conference of the North  
Graph convolutional networks (GCNs) are applied to these graphs and trained to perform multi-step reasoning.  ...  We introduce a neural model which integrates and reasons relying on information spread within documents and across multiple documents. We frame it as an inference problem on a graph.  ...  They jointly train graph neural networks and recurrent encoders.  ... 
doi:10.18653/v1/n19-1240 dblp:conf/naacl/CaoAT19 fatcat:7jdqxkfhgnetdcc5mjtlmji4xy

Personalized Graph Neural Networks with Attention Mechanism for Session-Aware Recommendation [article]

Shu Wu, Mengqi Zhang, Xin Jiang, Ke Xu, Liang Wang
2020 arXiv   pre-print
traditional Graph Neural Network (GNN) model, which considers the role of the user when the node embeddding is updated.  ...  To this end, we propose a novel method, named Personalized Graph Neural Networks with Attention Mechanism (A-PGNN) for brevity.  ...  For future work, we will improve the flexibility and scalability of PGNN by incorporating the dynamic graph neural networks.  ... 
arXiv:1910.08887v3 fatcat:jkkiqvthtbghlpaqlq7556crte

Column Networks for Collective Classification [article]

Trang Pham, Truyen Tran, Dinh Phung, Svetha Venkatesh
2016 arXiv   pre-print
An important task is collective classification, which is to jointly classify networked objects.  ...  We present Column Network (CLN), a novel deep learning model for collective classification in multi-relational domains.  ...  (via recurrent networks) (LeCun, Bengio, and Hinton 2015; Schmidhuber 2015) .  ... 
arXiv:1609.04508v2 fatcat:d74h6i6qungbjn5v5ncxb5eqla
« Previous Showing results 1 — 15 out of 1,221 results