Filters








105 Hits in 1.6 sec

Graph Neural Networks with Heterophily [article]

Jiong Zhu, Ryan A. Rossi, Anup Rao, Tung Mai, Nedim Lipka, Nesreen K. Ahmed, Danai Koutra
<span title="2021-06-14">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Graph Neural Networks (GNNs) have proven to be useful for many different practical applications.  ...  In this work, we propose a novel framework called CPGNN that generalizes GNNs for graphs with either homophily or heterophily.  ...  We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Quadro P6000 GPU used for this research.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2009.13566v3">arXiv:2009.13566v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/cwiivtjlxnck7bslu6fcvtx6oq">fatcat:cwiivtjlxnck7bslu6fcvtx6oq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210617014405/https://arxiv.org/pdf/2009.13566v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c9/f1/c9f1f39ac06ed03ccbbfb6bb04c86a45d26fe6b9.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2009.13566v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Graph Neural Networks for Graphs with Heterophily: A Survey [article]

Xin Zheng, Yixin Liu, Shirui Pan, Miao Zhang, Di Jin, Philip S. Yu
<span title="2022-02-14">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Recent years have witnessed fast developments of graph neural networks (GNNs) that have benefited myriads of graph analytic tasks and applications.  ...  However, as a ubiquitous graph property in numerous real-world scenarios, heterophily, i.e., nodes with different labels tend to be linked, significantly limits the performance of tailor-made homophilic  ...  The node features are represented by a feature matrix X ∈ Graph Neural Networks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.07082v1">arXiv:2202.07082v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/26d5mpnfzbb7fmloytdbj67yxa">fatcat:26d5mpnfzbb7fmloytdbj67yxa</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220217114752/https://arxiv.org/pdf/2202.07082v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/02/a3/02a3452a5f7fe42ba32bbf30af28b7845b2d6857.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2202.07082v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

On the Relationship between Heterophily and Robustness of Graph Neural Networks [article]

Jiong Zhu, Junchen Jin, Donald Loveland, Michael T. Schaub, Danai Koutra
<span title="2021-10-20">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Empirical studies on the robustness of graph neural networks (GNNs) have suggested a relation between the vulnerabilities of GNNs to adversarial attacks and the increased presence of heterophily in perturbed  ...  We theoretically and empirically show that for graphs exhibiting homophily (low heterophily), impactful structural attacks always lead to increased levels of heterophily, while for graph with heterophily  ...  We gratefully acknowledge the support of NVIDIA Corporation with  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.07767v2">arXiv:2106.07767v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/y2j3ox4ibbar3bl6coiesry32a">fatcat:y2j3ox4ibbar3bl6coiesry32a</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211026003407/https://arxiv.org/pdf/2106.07767v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/4a/4f/4a4f7e7a65d991d76e3d88dd2f4891de5c803d2e.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.07767v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

GBK-GNN: Gated Bi-Kernel Graph Neural Networks for Modeling Both Homophily and Heterophily [article]

Lun Du, Xiaozhou Shi, Qiang Fu, Xiaojun Ma, Hengyu Liu, Shi Han, Dongmei Zhang
<span title="2022-04-24">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Graph Neural Networks (GNNs) are widely used on a variety of graph-based machine learning tasks.  ...  We conduct extensive experiments on various datasets with different homophily-heterophily properties.  ...  GATED BI-KERNEL GRAPH NEURAL NETWORKS According to the identified problem in GCNs, we propose a new model pertinently, namely Gated Bi-Kernel Graph Neural Networks (GBK-GNN).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.15777v3">arXiv:2110.15777v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/vy3fql3avjbavoq3bkcoucg7ba">fatcat:vy3fql3avjbavoq3bkcoucg7ba</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220211223242/https://arxiv.org/pdf/2110.15777v2.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/12/15/1215a31dded88f5003bb165f3a2cc1016f1f6299.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.15777v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Improving Graph Neural Networks with Simple Architecture Design [article]

Sunil Kumar Maurya, Xin Liu, Tsuyoshi Murata
<span title="2021-05-17">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In this work, we decouple the node feature aggregation step and depth of graph neural network and introduce several key design strategies for graph neural networks.  ...  Graph Neural Networks have emerged as a useful tool to learn on the data by applying additional constraints based on the graph structure.  ...  Graph Neural Networks Graph Neural Networks (GNNs) leverage feature propagation mechanism [10] to aggregate neighborhood information of a node and use non-linear transformation with trainable weight  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2105.07634v1">arXiv:2105.07634v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/q4lxkvn2ffa75cwox24lrqndte">fatcat:q4lxkvn2ffa75cwox24lrqndte</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210519155142/https://arxiv.org/pdf/2105.07634v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/8e/6f/8e6f0800badceddde92ca062a4f6b107ee6ab744.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2105.07634v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Beyond Homophily in Graph Neural Networks

Danai Koutra
<span title="2022-04-22">2022</span> <i title="Zenodo"> Zenodo </i> &nbsp;
Invited talk in the graph deep learning track.  ...  Graph neural networks (GNNs).  ...  Graph Markov Neural Networks [21] model the joint label distribution with a conditional random field, trained with expectation maximization using GNNs. Table 2 : Design Comparison.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.5281/zenodo.6478758">doi:10.5281/zenodo.6478758</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/4klduazd45hhdovmz5i3hxy2gm">fatcat:4klduazd45hhdovmz5i3hxy2gm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220427050729/https://zenodo.org/record/6478759/files/22-KGC-v2-toshare-Koutra.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/74/71/74718a4f5a93487981c4aebd07315b8fc19c9c6c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.5281/zenodo.6478758"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="unlock alternate icon" style="background-color: #fb971f;"></i> zenodo.org </button> </a>

Simplifying approach to Node Classification in Graph Neural Networks [article]

Sunil Kumar Maurya, Xin Liu, Tsuyoshi Murata
<span title="2021-11-12">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Graph Neural Networks have become one of the indispensable tools to learn from graph-structured data, and their usefulness has been shown in wide variety of tasks.  ...  In this work, we decouple the node feature aggregation step and depth of graph neural network, and empirically analyze how different aggregated features play a role in prediction performance.  ...  Feature Selection Graph Neural Network Combining the model designs formulated earlier, we propose a simple and shallow (2-layered) graph GNN model called Feature Selection Graph Neural Network (FSGNN).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.06748v1">arXiv:2111.06748v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wswnvl3nonbmhi6jm2sn3cox3u">fatcat:wswnvl3nonbmhi6jm2sn3cox3u</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211121081037/https://arxiv.org/pdf/2111.06748v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/41/56/4156df84bf8b352f52b39304e4935a4994b1c150.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2111.06748v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Graph Pointer Neural Networks [article]

Tianmeng Yang, Yujing Wang, Zhihan Yue, Yaming Yang, Yunhai Tong, Jing Bai
<span title="2022-01-03">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Graph Neural Networks (GNNs) have shown advantages in various graph-based applications.  ...  In this work, we present Graph Pointer Neural Networks (GPNN) to tackle the challenges mentioned above.  ...  Related Work Graph Neural Networks Graph Neural Networks have many variants and applications.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.00973v2">arXiv:2110.00973v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/el2htkjkc5d33jsjnmneb2o5mq">fatcat:el2htkjkc5d33jsjnmneb2o5mq</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211006031057/https://arxiv.org/pdf/2110.00973v1.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/a4/62/a462735dcb112f60b091b8813d7793d1ff64b4a2.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2110.00973v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Learning heterophilious edge to drop: A general framework for boosting graph neural networks [article]

Jincheng Huang, Ping Li, Rui Huang, Chen Na
<span title="2022-05-23">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Graph Neural Networks (GNNs) aim at integrating node contents with graph structure to learn nodes/graph representations.  ...  Specifically, on assumption that graph smoothing along heterophilious edges can hurt prediction performance, we propose a structure learning method called LHE to identify heterophilious edges to drop.  ...  To allow the model to be aware of the information not only about node features but about class labels, graph markov neural networks [18] models the joint label distribution with conditional random field  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.11322v1">arXiv:2205.11322v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/rnyyopl5zbfllmlg56sxd7l4dy">fatcat:rnyyopl5zbfllmlg56sxd7l4dy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220525113855/https://arxiv.org/pdf/2205.11322v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/90/3a/903ac742e096162a4d7e2ae2b7a22e868079b8b6.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.11322v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

HL-Net: Heterophily Learning Network for Scene Graph Generation [article]

Xin Lin, Changxing Ding, Yibing Zhan, Zijian Li, Dacheng Tao
<span title="2022-05-04">2022</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Current SGG methods typically utilize graph neural networks (GNNs) to acquire context information between objects/relationships.  ...  Accordingly, in this paper, we propose a novel Heterophily Learning Network (HL-Net) to comprehensively explore the homophily and heterophily between objects/relationships in scene graphs.  ...  These methods can be roughly divided into two categories: Recurrent Neural Network (RNN)-based methods and Graph Neural Network (GNN)-based methods.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.01316v2">arXiv:2205.01316v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/syrkje7wszdynespd7a4rqsxai">fatcat:syrkje7wszdynespd7a4rqsxai</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220506085538/https://arxiv.org/pdf/2205.01316v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/d3/72/d372bc060c139c7f4478adb26c9e98858875c972.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2205.01316v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs [article]

Jiong Zhu, Yujun Yan, Lingxiao Zhao, Mark Heimann, Leman Akoglu, Danai Koutra
<span title="2020-10-23">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
We investigate the representation power of graph neural networks in the semi-supervised node classification task under heterophily or low homophily, i.e., in networks where connected nodes may have different  ...  and real networks with heterophily, respectively, and yield competitive performance under homophily.  ...  We gratefully acknowledge the support of NVIDIA Corporation with the donation of the Quadro P6000 GPU used for this research.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.11468v2">arXiv:2006.11468v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/jcu5qvoffnca3jxkw7obrbcxuy">fatcat:jcu5qvoffnca3jxkw7obrbcxuy</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201029020149/https://arxiv.org/pdf/2006.11468v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e0/f5/e0f550fa989769d3804518c525750c0a09857bc3.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.11468v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

GNNGuard: Defending Graph Neural Networks against Adversarial Attacks [article]

Xiang Zhang, Marinka Zitnik
<span title="2020-10-28">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
However, recent findings indicate that small, unnoticeable perturbations of graph structure can catastrophically reduce performance of even the strongest and most popular Graph Neural Networks (GNNs).  ...  The revised edges allow for robust propagation of neural messages in the underlying GNN.  ...  Background on graph neural networks.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.08149v3">arXiv:2006.08149v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/xkwcnoezwffgvp642hfcixjqwa">fatcat:xkwcnoezwffgvp642hfcixjqwa</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201030203129/https://arxiv.org/pdf/2006.08149v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/0e/7f/0e7fbbbb969ad8a4c76c78c31d181cf96d4e0ad4.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2006.08149v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Unifying Homophily and Heterophily Network Transformation via Motifs [article]

Yan Ge, Jun Ma, Li Zhang, Haiping Lu
<span title="2020-12-27">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Specifically, H2NT utilises motif representations to transform a network into a new network with a hybrid assumption via micro-level and macro-level walk paths.  ...  Because H2NT can sparsify networks with motif structures, it can also improve the computational efficiency of existing network embedding methods when integrated.  ...  The reason could be matrix factorisation and convolutional neural network are less sensitive to heterophily. Structural Role Classification.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.11400v2">arXiv:2012.11400v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/llwy5qsp6beuhf57d5gojadxbm">fatcat:llwy5qsp6beuhf57d5gojadxbm</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20201231055024/https://arxiv.org/pdf/2012.11400v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/2b/77/2b7729b92481af435b52c65af7e84db2be29bcd0.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2012.11400v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Tree Decomposed Graph Neural Network [article]

Yu Wang, Tyler Derr
<span title="2021-08-25">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
deeper graph neural networks.  ...  Moreover, we characterize the multi-hop dependency via graph diffusion within our tree decomposition formulation to construct Tree Decomposed Graph Neural Network (TDGNN), which can flexibly incorporate  ...  them together and present our Tree Decomposed Graph Neural Network (TDGNN).  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.11022v1">arXiv:2108.11022v1</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/5vnekq7t7zh2bkdrulg2zfgq6m">fatcat:5vnekq7t7zh2bkdrulg2zfgq6m</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210906084152/https://arxiv.org/pdf/2108.11022v1.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/c2/00/c2006bab73ecb2fb0f880786db34afe7277e3629.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2108.11022v1" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Is Homophily a Necessity for Graph Neural Networks? [article]

Yao Ma, Xiaorui Liu, Neil Shah, Jiliang Tang
<span title="2021-10-18">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Graph neural networks (GNNs) have shown great prowess in learning representations suitable for numerous graph-based machine learning tasks.  ...  Recent works design new architectures to overcome such heterophily-related limitations, citing poor baseline performance and new architecture improvements on a few heterophilous graph benchmark datasets  ...  Graph Neural Networks Graph neural networks learn node representations by aggregating and transforming information over the graph structure.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.06134v3">arXiv:2106.06134v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/wj5b47neorfpjjm3sulgydrqye">fatcat:wj5b47neorfpjjm3sulgydrqye</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20211021055120/https://arxiv.org/pdf/2106.06134v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/9f/3d/9f3dd3d44bab4cedb5b8831deadd5d6dbc1b4237.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2106.06134v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>
&laquo; Previous Showing results 1 &mdash; 15 out of 105 results