Learning Discrete Structures for Graph Neural Networks | AISC
Subscribers:
22,300
Published on ● Video Link: https://www.youtube.com/watch?v=e9QVrkKj-dg
For slides and more information on the paper, visit https://aisc.a-i.science/events/2019-05-27
Discussion lead: Sia Rezaei
Motivation:
Graph networks rely on the graph structure existing before hand which is not always the case. Here the authors present a novel method for learning the graph structure (using bilevel programming) simultaneously with training a GCN on the resulting graph.