AI-powered
podcast player
Listen to all your favourite podcasts with AI-powered features
3min chapter
#71 - ZAK JOST (Graph Neural Networks + Geometric DL) [UNPLUGGED]
Machine Learning Street Talk (MLST)
The Paradox of Machine Learning
If you have too much bias in your model is well witked, it can actually make you worse on average. This is getting to why we need to have several equiverant layerspermutation,. You would actually need to effectively perform message passing a discreet number of times in order to detect that structure effectively rit the number of gandinleers. There are some architectures that do things differently but ingeneral, one g an inlayer is one stage of message passing. And if the information you need is a five or six steps away, then you need five or six ganinlayers if you're just following the kind of standard approach.
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