Graph Neural Networks beyond Weisfeiler-Lehman and vanilla Message Passing

<p><em>This post is based on recent works with Cristian Bodnar, Xiaowen Dong, Ben Chamberlain, Davide Eynard, Fabrizio Frasca, Francesco Di Giovanni, Maria Gorinova, Pietro Li&ograve;, Giulia Luise, Sid Mishra, Guido Mont&uacute;far, Emanuele Rossi, Konstantin Rusch, Nina Otter, James Rowbottom, Jake Topping, Yu Guang Wang, and Stefan Webb. See also my previous post on&nbsp;Graph Neural Networks through the lens of Differential Geometry and Algebraic Topology.</em></p> <p>Graphs are a convenient way to abstract complex systems of relations and interactions. The increasing prominence of graph-structured data from social networks to high-energy physics to chemistry (all of these deal with objects that interact with each other, whether people, particles, or atoms), a series of&hellip;</p> <p><a href="https://towardsdatascience.com/graph-neural-networks-beyond-weisfeiler-lehman-and-vanilla-message-passing-bc8605fa59a"><strong>Click Here</strong></a></p>
Tags: Graph Neural