5min chapter

The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence) cover image

Learning Long-Time Dependencies with RNNs w/ Konstantin Rusch - #484

The TWIML AI Podcast (formerly This Week in Machine Learning & Artificial Intelligence)

CHAPTER

The Unicorn Paper - A Followon Paper to the Corn Paper

Instead of using coupled oscillators, we first uncouple them. So then you don't have any interconnection between the different nurans or hiden urants. And because of this kind of independency, and we're undamned, so we don't have some damping term in there any more, we can show that this system is actually a hemiltonian system. We are actually continues time hemitonian. The nice feature of hemitonian systems is that they are invertible in time. It's very memory efficient. And on top, as i said before, we don't have to train it on the same computer. That makes a lot of sense if

00:00

Get the Snipd
podcast app

Unlock the knowledge in podcasts with the podcast player of the future.
App store bannerPlay store banner

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

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