AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
4min chapter
#85 Dr. Petar Veličković (Deepmind) - Categories, Graphs, Reasoning [NEURIPS22 UNPLUGGED]
Machine Learning Street Talk (MLST)
Graph Neural Networks That Are Capable of Executing Algorithmic Computation
I think of algorithms as a sort of basic foundational building block of reasoning and it's kind of a timeless principle. We want to find ways inductive biases or training procedures to build neural networks that are more algorithmically capable in algorithmic reasoning. I'll give an example from Google Maps this is an application that I worked on at DeepMind so it's like something that I've thought about quite a bit you know we've invented these algorithms like Dijkstra's algorithm to be able to resolve these kinds of real-world routing problems.
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