Filters








88 Hits in 3.6 sec

End-to-End Learning from Complex Multigraphs with Latent-Graph Convolutional Networks [article]

Floris Hermsen, Peter Bloem, Fabian Jansen, Wolf Vos
2021 arXiv   pre-print
We study the problem of end-to-end learning from complex multigraphs with potentially very large numbers of edges between two vertices, each edge labeled with rich information.  ...  We propose Latent-Graph Convolutional Networks (L-GCNs), which propagate information from these complex edges to a latent adjacency tensor, after which further downstream tasks can be performed, such as  ...  In this study, we have shown that we can perform endto-end learning on complex multigraphs in two distinctively different settings, proposing Latent-Graph Convolutional Networks (L-GCNs).  ... 
arXiv:1908.05365v2 fatcat:wnofqwgpmfc5vd6l7yjxsdd2g4

Latent Patient Network Learning for Automatic Diagnosis [article]

Luca Cosmo, Anees Kazi, Seyed-Ahmad Ahmadi, Nassir Navab, Michael Bronstein
2020 arXiv   pre-print
Recently, Graph Convolutional Networks (GCNs) has proven to be a powerful machine learning tool for Computer Aided Diagnosis (CADx) and disease prediction.  ...  To this end, we propose a novel, end-to-end trainable graph learning architecture for dynamic and localized graph pruning.  ...  We show that using a single graph learned end-to-end allows both to achieve better performance and reduce the network complexity.  ... 
arXiv:2003.13620v1 fatcat:cdfxbi5sezaobil4xuc4xrnqum

Geometric Hawkes Processes with Graph Convolutional Recurrent Neural Networks

Jin Shang, Mingxuan Sun
2019 PROCEEDINGS OF THE THIRTIETH AAAI CONFERENCE ON ARTIFICIAL INTELLIGENCE AND THE TWENTY-EIGHTH INNOVATIVE APPLICATIONS OF ARTIFICIAL INTELLIGENCE CONFERENCE  
In this paper, we propose the Geometric Hawkes Process (GHP) model to better correlate individual processes, by integrating Hawkes processes and a graph convolutional recurrent neural network.  ...  The deep network structure is computational efficient since it requires constant parameters that are independent of the graph size.  ...  The authors would also like to thank Yichen Wang and Le Song from Georgia Tech for their helpful discussions.  ... 
doi:10.1609/aaai.v33i01.33014878 fatcat:3nbfznmb4ffnbpg2wd5qsp4ey4

Multi-Graph Convolutional-Recurrent Neural Network (MGC-RNN) for Short-Term Forecasting of Transit Passenger Flow [article]

Yuxin He, Lishuai Li, Xinting Zhu, Kwok Leung Tsui
2021 arXiv   pre-print
An innovative deep learning approach, Multi-Graph Convolutional-Recurrent Neural Network (MGC-RNN) is proposed to forecast passenger flow in urban rail transit systems to incorporate these complex factors  ...  The temporal dynamics of the inter-station correlations are also modeled via the proposed multi-graph convolutional-recurrent neural network structure.  ...  convolutional operators on multigraphs in our neural network model. 1) Multigraph Generation Graph generation is the fundamental step for GCNs.  ... 
arXiv:2107.13226v1 fatcat:isi7xdhkvjhozcfgnnwuxdz4ee

Graph Deep Learning: State of the Art and Challenges

S. Georgousis, M. P. Kenning, X. Xie
2021 IEEE Access  
Graphs can represent various complex systems, from molecular structure, to computer and social and traffic networks.  ...  The majority of GCNNs are designed to operate with certain properties. In this survey we review of the state of graph representation learning from the perspective of deep learning.  ...  [117] presented a model that learns from convolutions defined on edges with the goal of classifying human actions from skeleton graphs.  ... 
doi:10.1109/access.2021.3055280 fatcat:7ruskzkdkjgkfkia7drmww6lse

Brain Multigraph Prediction using Topology-Aware Adversarial Graph Neural Network [article]

Alaa Bessadok and Mohamed Ali Mahjoub and Islem Rekik
2021 arXiv   pre-print
The experimental results using five target domains demonstrated the outperformance of our method in brain multigraph prediction from a single graph in comparison with baseline approaches.  ...  Due to the high acquisition cost and processing time of multimodal MRI, existing deep learning frameworks based on Generative Adversarial Network (GAN) focused on predicting the missing multimodal medical  ...  graph attention network (GAT) and graph convolutional network (GCN)  ... 
arXiv:2105.02565v1 fatcat:q3pj6kh4wnfv3eunwu6cukunxu

Semi-supervised Classication for PolSAR Data with Multi-scale Evolving Weighted Graph Convolutional Network

Shijie Ren, Feng Zhou
2021 IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing  
To overcome this limitation and achieve robust PolSAR image classification, this article proposes the multiscale evolving weighted graph convolutional network, where weighted graphs based on superpixel  ...  In this article, we derive a new architectural design named graph evolving module that combines pairwise latent feature similarity and kernel diffusion to refine the graph structure in each scale.  ...  ACKNOWLEDGMENT The authors would like to thank the anonymous reviewers for their useful comments and constructive suggestions, which were of great help in improving this article, like to thank NASA/JPL  ... 
doi:10.1109/jstars.2021.3061418 fatcat:w2qo2zvoevc2zbxyte2kov766q

Spatial‐temporal attention wavenet: A deep learning framework for traffic prediction considering spatial‐temporal dependencies

Chenyu Tian, Wai Kin (Victor) Chan
2021 IET Intelligent Transport Systems  
Different from existing models, STAWnet does not need prior knowledge of the graph by developing a self-learned node embedding. These components are integrated into an end-to-end framework.  ...  Compared with traditional time-series and machine learning methods, deep learning models can flexibly handle relatively long time sequence and large traffic network structure.  ...  As a special case, dilated convolution with dilation 1 yields the standard convolution. In addition, a gated mechanism is used to learn complex temporal dependencies.  ... 
doi:10.1049/itr2.12044 fatcat:4fr5numrcjhulmtfa6pstwyxeu

Graph Representation Learning for Single Cell Biology

Leon Hetzel, David S. Fischer, Stephan Günnemann, Fabian J. Theis
2021 Current Opinion in Systems Biology  
Here, we discuss how graph representation learning maps to current models and concepts used in single-cell biology and formalise overlaps to developments in graphbased deep learning.  ...  Taking the inference of cell types or gene interactions as examples, graph representation learning has a wide applicability to both cell and gene graphs.  ...  Graph representation learning means to find a meaningful, potentially low-dimensional, representation of nodes from the complex relations present in a graph.  ... 
doi:10.1016/j.coisb.2021.05.008 fatcat:7kp5ignga5hibdduefw5xufkma

Graph Neural Networks in Network Neuroscience [article]

Alaa Bessadok, Mohamed Ali Mahjoub, Islem Rekik
2021 arXiv   pre-print
Relying on its non-Euclidean data type, graph neural network (GNN) provides a clever way of learning the deep graph structure and it is rapidly becoming the state-of-the-art leading to enhanced performance  ...  Here we review current GNN-based methods, highlighting the ways that they have been used in several applications related to brain graphs such as missing brain graph synthesis and disease classification  ...  Recent studies aimed to solve this issue by leveraging deep learning (DL) models such as convolutional neural network (CNN) which is inherently trained in an end-to-end fashion (36) .  ... 
arXiv:2106.03535v1 fatcat:jx7ixd7xjngthaq6qhb25gssm4

Learning on Attribute-Missing Graphs

Xu Chen, Siheng Chen, Jiangchao Yao, Huangjie Zheng, Ya Zhang, Ivor W. Tsang
2020 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Thereby, designing a new GNN for these graphs is a burning issue to the graph learning community.  ...  Graphs with complete node attributes have been widely explored recently.  ...  Federico et.al [10] proposed GCMC that combines a novel multigraph convolutional neural network and a recurrent neural network to better exploit the local stationary structures on the user-item graph  ... 
doi:10.1109/tpami.2020.3032189 pmid:33074805 fatcat:yx3qnk7gfzhxhowpvhcoexfxyi

Balanced Order Batching with Task-Oriented Graph Clustering [article]

Lu Duan, Haoyuan Hu, Zili Wu, Guozheng Li, Xinhang Zhang, Yu Gong, Yinghui Xu
2020 arXiv   pre-print
In this paper, rather than designing heuristics, we propose an end-to-end learning and optimization framework named Balanced Task-orientated Graph Clustering Network (BTOGCN) to solve the BOBP by reducing  ...  In BTOGCN, a task-oriented estimator network is introduced to guide the type-aware heterogeneous graph clustering networks to find a better clustering result related to the BOBP objective.  ...  an end-to-end learning and optimization framework BTOGCN to solve a very complex combinational optimization problem named BOBP.  ... 
arXiv:2008.09018v1 fatcat:5kxqjc5py5fmrbor6vx7sljvre

Explainable Link Prediction for Emerging Entities in Knowledge Graphs [article]

Rajarshi Bhowmik, Gerard de Melo
2020 arXiv   pre-print
To overcome this issue, we propose an inductive representation learning framework that is able to learn representations of previously unseen entities.  ...  Despite their large-scale coverage, cross-domain knowledge graphs invariably suffer from inherent incompleteness and sparsity.  ...  Acknowledgement We thank Diffbot for their grant support to Rajarshi Bhowmik's work. We also thank Diffbot and Google for providing the computing infrastructure required for this project.  ... 
arXiv:2005.00637v2 fatcat:s2mkqqwbczeadg4qyi2os2ezj4

Utilising Graph Machine Learning within Drug Discovery and Development [article]

Thomas Gaudelet, Ben Day, Arian R. Jamasb, Jyothish Soman, Cristian Regep, Gertrude Liu, Jeremy B. R. Hayter, Richard Vickers, Charles Roberts, Jian Tang, David Roblin, Tom L. Blundell (+2 others)
2021 arXiv   pre-print
Graph Machine Learning (GML) is receiving growing interest within the pharmaceutical and biotechnology industries for its ability to model biomolecular structures, the functional relationships between  ...  learning.  ...  ACKNOWLEDGEMENTS We gratefully acknowledge support from William L. Hamilton, Benjamin Swerner, Lyuba V. Bozhilova, and Andrew Anighoro.  ... 
arXiv:2012.05716v2 fatcat:kre2kx3x4ff43mmuh7khrxmmzy

Two person Interaction Recognition Based on Effective Hybrid Learning

2019 KSII Transactions on Internet and Information Systems  
We broaden a semi-supervised learning method combined with an active learning method to improve overall performance.  ...  Our approach exploits a pre-trained network model (the VGG16 from the University of Oxford Visual Geometry Group) and extends the Faster R-CNN (region-based convolutional neural network a stateof-the-art  ...  Compare < −1 If < −1 Roll back to −1 End if End for End Experiments and Results The main objectives of our experiments are to prove the efficiency of the framework with different conditional data of  ... 
doi:10.3837/tiis.2019.02.015 fatcat:h7crokxajnfsnpnl564zk2ng7y
« Previous Showing results 1 — 15 out of 88 results