没有合适的资源?快使用搜索试试~ 我知道了~
首页Deeplearning with graph structured representations.pdf
Deeplearning with graph structured representations.pdf
需积分: 0 9 下载量 95 浏览量
更新于2023-05-13
评论
收藏 8.7MB PDF 举报
图卷积神经网络--Thomas Kipf 的PHD论文,涵盖GCN的基本理论以及在无监督学习的应用。
资源详情
资源评论
资源推荐
UvA-DARE is a service provided by the library of the University of Amsterdam (http://dare.uva.nl)
UvA-DARE (Digital Academic Repository)
Deep learning with graph-structured representations
Kipf, T.N.
Link to publication
Creative Commons License (see https://creativecommons.org/use-remix/cc-licenses):
Other
Citation for published version (APA):
Kipf, T. N. (2020). Deep learning with graph-structured representations.
General rights
It is not permitted to download or to forward/distribute the text or part of it without the consent of the author(s) and/or copyright holder(s),
other than for strictly personal, individual use, unless the work is under an open content license (like Creative Commons).
Disclaimer/Complaints regulations
If you believe that digital publication of certain material infringes any of your rights or (privacy) interests, please let the Library know, stating
your reasons. In case of a legitimate complaint, the Library will make the material inaccessible and/or remove it from the website. Please Ask
the Library: https://uba.uva.nl/en/contact, or a letter to: Library of the University of Amsterdam, Secretariat, Singel 425, 1012 WP Amsterdam,
The Netherlands. You will be contacted as soon as possible.
Download date: 13 May 2020
Deep Learning with
Graph-Structured
Representations
Deep Learning with Graph-Structured Representations Thomas Kipf
Thomas Kipf
Deep Learning with
Graph-Structured Representations
ACADEMISCH PROEFSCHRIFT
ter verkrijging van de graad van doctor
aan de Universiteit van Amsterdam
op gezag van de Rector Magnificus
prof. dr. ir. K.I.J. Maex
ten overstaan van een door het College voor Promoties
ingestelde commissie,
in het openbaar te verdedigen
op donderdag 23 april 2020, te 14.00 uur
door
Thomas Norbert Kipf
geboren te Roth
P R O M O T I E C O M M I S S I E
Promotor:
prof. dr. M. Welling Universiteit van Amsterdam
Copromotor:
dr. I.A. Titov University of Edinburgh
Overige leden:
prof. dr. M.M. Bronstein Imperial College London
prof. dr. F.A.H. van Harmelen Vrije Universiteit Amsterdam
prof. dr. M. de Rijke Universiteit van Amsterdam
dr. P.W. Battaglia DeepMind Technologies Ltd
dr. H.C. van Hoof Universiteit van Amsterdam
Faculteit der Natuurwetenschappen, Wiskunde en Informatica
The work described in this thesis has been primarily carried out at the Ams-
terdam Machine Learning Lab (AMLab) of the University of Amsterdam and
in part during an internship at DeepMind Technologies Ltd, London, UK. The
research carried out at the University of Amsterdam was funded by SAP SE.
Printed by Ridderprint, The Netherlands.
ISBN: 978-94-6375-851-2
Copyright
c
2020 by T. N. Kipf, Amsterdam, The Netherlands.
ii
S U M M A R Y
In this thesis, Deep Learning with Graph-Structured Representations, we propose
novel approaches to machine learning with structured data. Our proposed
methods are largely based on the theme of structuring the representations and
computations of neural network-based models in the form of a graph, which
allows for improved generalization when learning from data with both explicit
and implicit modular structure.
Our contributions are as follows:
• We propose graph convolutional networks (GCNs) (Kipf and Welling,
2017; Chapter 3) for semi-supervised classification of nodes in graph-
structured data. GCNs are a form of graph neural network that per-
form parameterized message-passing operations in a graph, modeled as a
first-order approximation to spectral graph convolutions. GCNs achieved
state-of-the-art performance in node-level classification tasks in a number
of undirected graph datasets at the time of publication.
• We propose graph auto-encoders (GAEs) (Kipf and Welling, 2016; Chap-
ter 4) for unsupervised learning and link prediction in graph-structured
data. GAEs utilize an encoder based on graph neural networks and a
decoder that reconstructs links in a graph based on a pairwise scoring
function. We further propose a model variant framed as a probabilistic
generative model that is trained using variational inference, which we
name variational GAE. GAEs and variational GAEs are particularly suited
for representation learning on graphs in the absence of node labels.
• We propose relational GCNs (Schlichtkrull and Kipf et al., 2018; Chapter
5) that extend the GCN model to directed, relational graphs with multi-
ple edge types. Relational GCNs are well-suited for modeling relational
data and we demonstrate an application to semi-supervised entity classi-
fication in knowledge bases.
iii
剩余177页未读,继续阅读
zybjtu
- 粉丝: 72
- 资源: 22
上传资源 快速赚钱
- 我的内容管理 收起
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
会员权益专享
最新资源
- RTL8188FU-Linux-v5.7.4.2-36687.20200602.tar(20765).gz
- c++校园超市商品信息管理系统课程设计说明书(含源代码) (2).pdf
- 建筑供配电系统相关课件.pptx
- 企业管理规章制度及管理模式.doc
- vb打开摄像头.doc
- 云计算-可信计算中认证协议改进方案.pdf
- [详细完整版]单片机编程4.ppt
- c语言常用算法.pdf
- c++经典程序代码大全.pdf
- 单片机数字时钟资料.doc
- 11项目管理前沿1.0.pptx
- 基于ssm的“魅力”繁峙宣传网站的设计与实现论文.doc
- 智慧交通综合解决方案.pptx
- 建筑防潮设计-PowerPointPresentati.pptx
- SPC统计过程控制程序.pptx
- SPC统计方法基础知识.pptx
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功
评论0