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)
The Importance of Categories in Geometric Deep Learning
Categorical concepts like categories, functors and natural transformations basically generalize all the stuff of geometric deep learning into the wrong beyond. In our paper we tried to use exactly these kinds of category theoretic tools to study what it would mean to build a graph neural network that is capable of behaving like a classical computer science algorithm. But one thing that people very often do when they want to predict not only the outputs but also in the edges is to implement GNNs with categorical representations. And just because you cannot implement it mathematically doesn't mean this is no longer a problem from the categorical perspective. It gives us a very interesting sort of if you've done any functional programming a
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