Deep learning for graph structured data

Graph-structured data is ubiquitous across diverse domains, representing valuable relational information between entities. However, most deep learning techniques like convolutional and recurrent neural networks are tailored for grid-structured data and struggle to handle such graphs. This has led to...

Full description

Saved in:
Bibliographic Details
Main Author: Dwivedi Vijay Prakash
Other Authors: Luu Anh Tuan
Format: Thesis-Doctor of Philosophy
Language:English
Published: Nanyang Technological University 2024
Subjects:
Online Access:https://hdl.handle.net/10356/175787
Tags: Add Tag
No Tags, Be the first to tag this record!
Institution: Nanyang Technological University
Language: English
id sg-ntu-dr.10356-175787
record_format dspace
spelling sg-ntu-dr.10356-1757872024-06-03T06:51:19Z Deep learning for graph structured data Dwivedi Vijay Prakash Luu Anh Tuan School of Computer Science and Engineering anhtuan.luu@ntu.edu.sg Computer and Information Science Deep learning Graphs Graph representation learning Graph neural network Graph-structured data is ubiquitous across diverse domains, representing valuable relational information between entities. However, most deep learning techniques like convolutional and recurrent neural networks are tailored for grid-structured data and struggle to handle such graphs. This has led to growing interest in graph representation learning using graph neural networks (GNNs). GNNs integrate graph structure into neural network layers through message-passing in general. However, several challenges exist like the lack of rigorous benchmarks, limitations in model expressiveness, and poor scalability. This thesis aims to advance graph representation learning by tackling these key challenges. First, it develops comprehensive benchmarks for standardized assessment of GNNs. This includes medium-scale tasks covering supervised and semi-supervised node, edge and graph classifications across domains like social networks, computer vision, and combinatorial optimization. The thesis also introduces a novel benchmark specifically designed to test modeling of long-range interactions in larger graphs. Second, the thesis focuses on developing new GNN architectures for learning on graphs with higher expressivity and generalization. It extends Transformer networks to graph domains by introducing graph-based inductive biases like leveraging sparsity and designing Laplacian positional encodings. Another technique learns separate structural and positional representations in GNNs through using informative graph diffusion features. This boosts model capacity significantly. Finally, the thesis addresses the problem of scaling graph models, in particular Graph Transformers, to massive graphs. It investigates design principles like incorporating efficient local and global graph representations. Following this, a scalable Graph Transformer framework is proposed. It uses novel neighborhood sampling and global attention schemes to capture both local structure and global dependencies in very large graphs. Overall, through rigorous benchmarks, expressive architectures, and scalable models, this thesis makes significant contributions towards advancing deep learning on graph- structured data across multiple fronts. The techniques pave the way for adoption of GNNs in real-world applications involving complex relational data. Doctor of Philosophy 2024-05-21T08:18:17Z 2024-05-21T08:18:17Z 2024 Thesis-Doctor of Philosophy Dwivedi Vijay Prakash (2024). Deep learning for graph structured data. Doctoral thesis, Nanyang Technological University, Singapore. https://hdl.handle.net/10356/175787 https://hdl.handle.net/10356/175787 10.32657/10356/175787 en This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License (CC BY-NC 4.0). application/pdf Nanyang Technological University
institution Nanyang Technological University
building NTU Library
continent Asia
country Singapore
Singapore
content_provider NTU Library
collection DR-NTU
language English
topic Computer and Information Science
Deep learning
Graphs
Graph representation learning
Graph neural network
spellingShingle Computer and Information Science
Deep learning
Graphs
Graph representation learning
Graph neural network
Dwivedi Vijay Prakash
Deep learning for graph structured data
description Graph-structured data is ubiquitous across diverse domains, representing valuable relational information between entities. However, most deep learning techniques like convolutional and recurrent neural networks are tailored for grid-structured data and struggle to handle such graphs. This has led to growing interest in graph representation learning using graph neural networks (GNNs). GNNs integrate graph structure into neural network layers through message-passing in general. However, several challenges exist like the lack of rigorous benchmarks, limitations in model expressiveness, and poor scalability. This thesis aims to advance graph representation learning by tackling these key challenges. First, it develops comprehensive benchmarks for standardized assessment of GNNs. This includes medium-scale tasks covering supervised and semi-supervised node, edge and graph classifications across domains like social networks, computer vision, and combinatorial optimization. The thesis also introduces a novel benchmark specifically designed to test modeling of long-range interactions in larger graphs. Second, the thesis focuses on developing new GNN architectures for learning on graphs with higher expressivity and generalization. It extends Transformer networks to graph domains by introducing graph-based inductive biases like leveraging sparsity and designing Laplacian positional encodings. Another technique learns separate structural and positional representations in GNNs through using informative graph diffusion features. This boosts model capacity significantly. Finally, the thesis addresses the problem of scaling graph models, in particular Graph Transformers, to massive graphs. It investigates design principles like incorporating efficient local and global graph representations. Following this, a scalable Graph Transformer framework is proposed. It uses novel neighborhood sampling and global attention schemes to capture both local structure and global dependencies in very large graphs. Overall, through rigorous benchmarks, expressive architectures, and scalable models, this thesis makes significant contributions towards advancing deep learning on graph- structured data across multiple fronts. The techniques pave the way for adoption of GNNs in real-world applications involving complex relational data.
author2 Luu Anh Tuan
author_facet Luu Anh Tuan
Dwivedi Vijay Prakash
format Thesis-Doctor of Philosophy
author Dwivedi Vijay Prakash
author_sort Dwivedi Vijay Prakash
title Deep learning for graph structured data
title_short Deep learning for graph structured data
title_full Deep learning for graph structured data
title_fullStr Deep learning for graph structured data
title_full_unstemmed Deep learning for graph structured data
title_sort deep learning for graph structured data
publisher Nanyang Technological University
publishDate 2024
url https://hdl.handle.net/10356/175787
_version_ 1800916304280944640