Data Skeptic cover image

Long Term Time Series Forecasting

Data Skeptic

00:00

Is There a Parallel to Support Vector Machines?

The kernel method takes a vector of signs with different frequencies and warfs them into some non linear, quasi periodic functions. This sort of reflects how we have these perfectly linear systems such as the earth's orbit of the sun and the earth's rotation. These are kind of small models because they overfit very quickly. But if you have a problem that is highly nondina than you would mornin moniarity, it's kangf likeThese copa models, they're almost like on ra well, just keep on doing what they're doing. Well, ie, wax, like Wax? So surprisingly surprisingly good at finding things. Because it is stablee.

Transcript
Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!
App store bannerPlay store banner
Get the app