Adaptive Non-Linear Control, Rob Sannar on Neural Networks

The key aspect for us was that you had lots of parameters. And so if we're trying to do things systematically with lots of parameters doing this mathematical expansion was very natural thing, right? But that also got me to then talk to other people, some of which became some of my best friends like Zith and Mala to try to be to say, well, you know, some of Gaussians were nice. So this is probably an even better mathematical way to do things, perhaps wavelets or perhaps other things, you know.

Transcript

Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app