网络嵌入和图卷积神经网络技术概述_第1页
网络嵌入和图卷积神经网络技术概述_第2页
网络嵌入和图卷积神经网络技术概述_第3页
网络嵌入和图卷积神经网络技术概述_第4页
网络嵌入和图卷积神经网络技术概述_第5页
已阅读5页,还剩27页未读 继续免费阅读

下载本文档

版权说明:本文档由用户提供并上传,收益归属内容提供方,若内容存在侵权,请进行举报或认领

文档简介

1、网络嵌入和图卷积神经网络技术概述技术创新 变革未来网络/图数据图是对于数据的一/通用、全面、复杂的表示形式网络无处不在社交网络生物网络金融网络物联网信息网络物流网络为什么网络很重要?我们很少只关心数据本身,而不关心数据之间的关联Reflected by relational subjectsDecided by relational subjectsTargetImage CharacterizationTargetSocial Capital网络数据对机器学习模型 不友好G = ( V, E )LinksTopologyInapplicability of ML methodsNetwork

2、 DataFeature ExtractionPattern DiscoveryPipeline for network analysisNetwork ApplicationsLearnabilityLearning from NetworksNetwork EmbeddingGNNG = ( V, E )G = ( V )Vector SpacegenerateembedEasy to parallelCan apply classical ML methods网络嵌入 (Network Embedding)网络嵌入的目标GoalSupport network inference in v

3、ector spaceReflect network structureMaintain network propertiesBACTransitivitypBasic idea: recursive definition of statespA simple example: PageRank图神经网络GNNF. Scarselli, et al. The graph neural network model. IEEE TNN, 2009.定义在图拓扑上的学习框架pMain idea: pass messages between pairs of nodes & agglomeratepS

4、tacking multiple layers like standard CNNs:pState-of-the-art results on node classification图卷积神经网络GCNT. N. Kipf and M. Welling. Semi-supervised classification with graph convolutional networks. ICLR, 2017.图神经网络GNN简史网络嵌入与图神经网络GraphFeatureNetwork EmbeddingInputModelOutputEmbeddingTask resultsFeatureTo

5、pology to VectorGCNTask resultsFusion of Topology and FeaturesUnsupervised vs. (Semi-)Supervised图卷积网络 v . 网络嵌入pIn some sense, they are different.pGraphs exist in mathematics. (Data Structure)p Mathematical structures used to model pairwise relations between objectspNetworks exist in the real world.

6、(Data)p Social networks, logistic networks, biology networks, transaction networks, etc.pA network can be represented by a graph.pA dataset that is not a network can also be represented by a graph.图卷积网络应用于自然语言处理p Many papers on BERT + GNN.p BERT is for retrieval.pIt creates an initial graph of relev

7、ant entities and the initial evidence.p GNN is for reasoning.pIt collects evidence (i.e., old messages on the entities) and arrive at new conclusions (i.e., new messages on the entities), by passing the messages around and aggregating them.Cognitive Graph for Multi-Hop Reading Comprehension at Scale

8、. Ding et al., ACL 2019.Dynamically Fused Graph Network for Multi-hop Reasoning. Xiao et al., ACL 2019.图卷积网络应用于计算机视觉pA popular trend in CV is to construct a graph during the learning process.pTo process multiple objects or parts in a scene, and to infer their relationships.pExample: Scene graphs.Sce

9、ne Graph Generation by Iterative Message Passing. Xu et al., CVPR 2017.Image Generation from Scene Graphs. Johnson et al., CVPR 2018.图卷积网络应用于符号推理pWe can view the process of symbolic reasoning as a directed acyclic graph.pMany recent efforts use GNNs to perform symbolic reasoning.Learning by Abstract

10、ion: The Neural State Machine. Hudson & Manning, 2019.Can Graph Neural Networks Help Logic Reasoning? Zhang et al., 2019.Symbolic Graph Reasoning Meets Convolutions. Liang et al., NeurIPS 2018.pStructural equation modeling, a form of causal modeling, tries to describe the relationships between the v

11、ariables as a directed acyclic graph (DAG).pGNN can be used to represent a nonlinear structural equation and help find the DAG, after treating the adjacency matrix as parameters.图卷积网络应用于结构方程建模DAG-GNN: DAG Structure Learning with Graph Neural Networks. Yu et al., ICML 2019.(大多数)图卷积网络方法的Pipeline Co-oc

12、currence (neighborhood)网络嵌 : 拓扑向量化 High-order proximities网络嵌 : 拓扑向量化 Communities网络嵌 : 拓扑向量化 Heterogeneous networks网络嵌 : 拓扑向量化(大多数)网络嵌入方法的PipelineLearning for Networks v.s. Learning via GraphsLearning for networksLearning Via GraphsNetwork EmbeddingGCN网络嵌入方法解决的核心问题Node NeighborhoodCommunityPair-wise

13、ProximityHyper EdgesGlobal StructureReducing representation dimensionality while preserving necessary topologicalstructures and properties.Nodes & LinksNon-transitivityAsymmetric TransitivityDynamicUncertaintyHeterogeneityInterpretabilityTopology-driven图卷积神经网络方法解决的核心问题Fusing topologyand featuresin t

14、he wayof smoothing features with the assistance of topology.Feature-driven如果问题是拓扑驱动的?Since GCN is filtering features, it is inevitably feature-drivenStructure only provides auxiliary information (e.g. for filtering/smoothing)When feature plays the key role, GNN performs good How about the contrary?S

15、ynthesis data: stochastic block model + random featuresMethodResultsRandom10.0GCN18.31.1DeepWalk99.00.1网络嵌入 vs 图神经网络There is no better one, but there is more proper one.反思:图神经网络是否真的是深度学习方法? This simplified GNN (SGC) showsremarkable results:Node classificationText Classification反思:图神经网络是否真的是深度学习方法?Wu, Felix, et al. S

温馨提示

  • 1. 本站所有资源如无特殊说明,都需要本地电脑安装OFFICE2007和PDF阅读器。图纸软件为CAD,CAXA,PROE,UG,SolidWorks等.压缩文件请下载最新的WinRAR软件解压。
  • 2. 本站的文档不包含任何第三方提供的附件图纸等,如果需要附件,请联系上传者。文件的所有权益归上传用户所有。
  • 3. 本站RAR压缩包中若带图纸,网页内容里面会有图纸预览,若没有图纸预览就没有图纸。
  • 4. 未经权益所有人同意不得将文件中的内容挪作商业或盈利用途。
  • 5. 人人文库网仅提供信息存储空间,仅对用户上传内容的表现方式做保护处理,对用户上传分享的文档内容本身不做任何修改或编辑,并不能对任何下载内容负责。
  • 6. 下载文件中如有侵权或不适当内容,请与我们联系,我们立即纠正。
  • 7. 本站不保证下载资源的准确性、安全性和完整性, 同时也不承担用户因使用这些下载资源对自己和他人造成任何形式的伤害或损失。

评论

0/150

提交评论