We recently conducted a small R&D project using a fast.ai language model for graph classification and compared accuracy to a recent paper which uses a CNN. I wrote a short blog yesterday about our results which has a link to our repo in github, if interested. In short, we passed in random walks over graph, enriched with structural and graph property information, to an RNN and trained a classifier with the purpose of labeling small graphs. Thank you Jeremy and the community for fast.ai!
5 Likes
That is REALLY interesting and creative work - I’m going to share with my team right now, as I think they’ll all be excited to read about it. Thanks for sharing - it’s one of the more novel approaches to a problem I’ve seen in a while.
1 Like
Thank you Matthew. Glad that our work might help you and your team!