Graph Neural Networks
Tim and Zach discuss the challenges of message passing in graph neural networks, highlighting issues like over-smoothing and over-squashing problems. They delve into the complexities of detecting structures in graphs and the need for multiple equivariant layers for effective computation.In this clip
From this podcast
Machine Learning Street Talk (MLST)
#71 - ZAK JOST (Graph Neural Networks + Geometric DL) [UNPLUGGED]
Related Questions