BEGIN:VCALENDAR
PRODID:-//Columba Systems Ltd//NONSGML CPNG/SpringViewer/ICal Output/3.3-
M3//EN
VERSION:2.0
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VEVENT
DTSTAMP:20231016T104741Z
DTSTART:20231027T124500Z
DTEND:20231027T143000Z
SUMMARY:IDSAI Seminar: Petar Velickovic
UID:{http://www.columbasystems.com/customers/uom/gpp/eventid/}r231-lnlnju
n7-kxjyt5
DESCRIPTION:Speaker: Petar Velickovic\, Staff Research Scientist\, Google
DeepMind and Affiliated Lectureship at the University of Cambridge\n\nA
bstract: When deploying graph neural networks\, we often make a seemingl
y innocent assumption: that the input graph we are given is the ground-t
ruth. However\, as my talk will unpack\, this is often not the case: eve
n when the graphs are perfectly correct\, they may be severely suboptima
l 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\, inc
luding two of our own contributions (https://arxiv.org/abs/2210.02997\,
https://arxiv.org/abs/2306.03589)\, one of which won the Best Paper Awar
d at the Graph Learning Frontiers Workshop at NeurIPS'22.\n\nPetar Velic
kovic is a Research Scientist at DeepMind. He holds a PhD degree from th
e University of Cambridge (obtained under the supervision of Pietro LiĆ²)
\, with prior collaborations at Nokia Bell Labs and Mila. His current re
search interests broadly involve devising neural network architectures t
hat operate on nontrivially structured data (such as graphs)\, and their
applications in algorithmic reasoning and computational biology.\n\nPet
ar has published his work in these areas at both machine learning venues
(ICLR\, NeurIPS-W\, ICML-W) and biomedical venues and journals (Bioinfo
rmatics\, PLOS One\, JCB\, PervasiveHealth). In particular\, he is the f
irst author of Graph Attention Networks\, a popular convolutional layer
for graphs\, and Deep Graph Infomax\, a scalable local/global unsupervis
ed 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.\n\nAgenda:\
n\n13:45 - Registration and Arrival\n\n14:00 - Welcome and introductions
\n\n14:10 - Petar Velickovic\, Staff Research Scientist\, Google DeepMin
d and Affiliated Lectureship at the University of Cambridge (Graph Neura
l Networks / Geometric deep learning.)\n\n14:45 - Q&A\n\n15:00 - Network
ing\n\n15:30 - Event close
STATUS:TENTATIVE
TRANSP:TRANSPARENT
CLASS:PUBLIC
END:VEVENT
END:VCALENDAR