AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
6min chapter
#85 Dr. Petar Veličković (Deepmind) - Categories, Graphs, Reasoning [NEURIPS22 UNPLUGGED]
Machine Learning Street Talk (MLST)
Using Expander Graphs in a Data Driven Way Is Necessary
We assume in many tasks like say molecular property prediction tasks you actually want data to travel globally so that's exactly what we do. We propose to propagate information over these expander graphs which are known constructs from graph theory specifically expander graphs have mathematical properties of a high-chigger constant and very low bottlenecks. The only generative parameter of these graphs is the size of the group this end over here. Our proposal take basically your state-of-the-art graph net that you care about and switch the graph neural network connectivity in every even layer to operate over one of these guys rather than the input graph.
00:00
Transcript
Episode notes
Get the Snipd
podcast app
Unlock the knowledge in podcasts with the podcast player of the future.
AI-powered
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode
Save any
moment
Hear something you like? Tap your headphones to save it with AI-generated key takeaways
Share
& Export
Send highlights to Twitter, WhatsApp or export them to Notion, Readwise & more
AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
Discover
highlights
Listen to the best highlights from the podcasts you love and dive into the full episode