Decoupling the input graph and the computational graph: The most important unsolved problem in graph representation learning
Abstract: When deploying graph neural networks, we often make a seemingly innocent assumption: that the input graph we are given is the ground-truth. However, as my talk will unpack, this is often not the case: even when the graphs are perfectly correct, they may be severely suboptimal for completing the task at hand. This will introduce us to a rich and vibrant area of graph rewiring, which is experiencing a renaissance in recent times. I will discuss some of the most representative works, including two of our own contributions (https://arxiv.org/abs/2210.02997, https://arxiv.org/abs/2306.03589), one of which won the Best Paper Award at the Graph Learning Frontiers Workshop at NeurIPS'22.
Petar Veličković, Staff Research Scientist, Google DeepMind and Affiliated Lectureship at the University of Cambridge
Petar Veličković is a Research Scientist at DeepMind. He holds a PhD degree from the University of Cambridge (obtained under the supervision of Pietro Liò), with prior collaborations at Nokia Bell Labs and Mila. His current research interests broadly involve devising neural network architectures that operate on nontrivially structured data (such as graphs), and their applications in algorithmic reasoning and computational biology.
Petar has published his work in these areas at both machine learning venues (ICLR, NeurIPS-W, ICML-W) and biomedical venues and journals (Bioinformatics, PLOS One, JCB, PervasiveHealth). In particular, he is the first author of Graph Attention Networks, a popular convolutional layer for graphs, and Deep Graph Infomax, a scalable local/global unsupervised learning pipeline for graphs. His research has been featured in media outlets such as ZDNet. Additionally, he has co-organised workshops on Graph Representation Learning at ICLR 2019 and NeurIPS 2019.
Agenda:
13:45 - Registration and Arrival
14:00 - Welcome and introductions with Dr Julia Handl, Professor of Decision Sciences at the University of Manchester
14:10 - Petar Veličković, Staff Research Scientist, Google DeepMind and Affiliated Lectureship at the University of Cambridge (Graph Neural Networks / Geometric deep learning.)
14:45 - Q&A
15:00 - Networking
15:30 - Event close
This seminar is hosted in partnership with ID Manchester and Turing Innovation Catalyst as part of the official AI Fringe.
The AI Fringe is a series of events hosted across London and the UK to complement the UK Government’s AI Safety Summit by bringing a broad and diverse range of voices into the conversation. It will expand discussion around safe and responsible AI beyond the AI Safety Summit’s focus on Frontier AI safety.
