Exploring Local Information for Graph Representation Learning

[CFAR Rising Star Lecture Series]
Exploring Local Information for Graph Representation Learning by Dr Zhang Li
30 Jan 2023 | 4.00pm (Singapore Time)

Graphs are important data structures that can capture interactions between individual entities. The primitive graph representation is usually high-dimensional, sparse and noisy, which is challenging for direct usage in downstream tasks (e.g., node or graph classification, link prediction). Various graph representation learning (GRL) techniques have been developed to convert the raw graph data into low-dimensional vector representations while preserving the intrinsic graph properties. Currently, graph neural networks (GNNs) are the most popular paradigm that could utilise nodes’ local information to assist their representation learning.
 
Local neighbourhood information in graphs varies greatly for different nodes. In this talk, Dr Zhang Li will introduce new GNNs by exploring and modelling different types of local information to tackle the weaknesses of current GNNs in both algorithms and applications. The three new GNN models are: 1) node feature convolution for graph convolutional network (NFC-GCN) to consider feature-level attention of local information; 2) learnable aggregator for GCN (LA-GCN) to generalise NFC-GCN further by lifting constraints on the input data format; and 3) hop-hop relation-aware GNN (HHR-GNN) to incorporate hop-level attention of local information. Lastly, Dr Zhang will present ways to apply GNNs to two industrial graphs for personalised video search and cross-domain recommendation tasks.


SPEAKER
talks---zhang-li
Dr Zhang Li
Postdoctoral Research Assistant 
PhD in Machine Learning Group
Department of Computer Science
University of Sheffield
Dr Zhang Li is a Postdoctoral Research Assistant working on machine learning for network analysis, a joint project between the Machine Learning Research Group and Oxford e-Research Centre at the University of Oxford. Before this, Dr Zhang got her PhD in the Machine Learning Group, Department of Computer Science at the University of Sheffield. Her research mainly focuses on network analysis, graph representation learning, graph neural networks, recommendation and search.