Efficient Training and Inference of Graph Neural Networks: A Literature Review

PhD Qualifying Examination


Title: "Efficient Training and Inference of Graph Neural Networks: A
Literature Review"

by

Mr. Qiqi ZHOU


Abstract:

Graph Neural Networks (GNNs) have achieved remarkable success in various
graph-related tasks across a wide range of domains. However, the rapid
growth of real-world graph data and GNN model complexity have exposed
fundamental efficiency challenges in both the training and inference of GNN
models. In recent years, numerous studies have proposed to address the
challenges from different perspectives. This survey summarizes and analyzes
the underlying causes of these efficiency issues and provides a structured
review of existing techniques for efficient GNN training and inference. This
work also explores promising research directions and open challenges in this
field.


Date:                   Friday, 28 November 2025

Time:                   10:00am - 12:00noon

Venue:                  Room 3494 (Lift 25/26)

Committee Members:      Prof. Lei Chen (Supervisor, Chairperson)
                        Dr. Junxian He
                        Dr. Lei Li
                        Prof. Qiong Luo