Learning Bayesian Statistics cover image

#37 Prophet, Time Series & Causal Inference, with Sean Taylor

Learning Bayesian Statistics

00:00

How Do You Know if You're Not Overfitting Ever?

Every day you get a new teteset to test your model. It's quite a different way of thinking about problems, exactly. And it also changes the way that you think about modelling as a process rather than a thing. A model isn't a snapshot. Any reasonable model tht yu're going to use reality is going to have to be capable of incorporating new data as time goes on. Yes, therere certainly, like, o argument to be made about the evolutionary dynamic. How often are your models subjected to some evolutionary pressure of being held accountable for further decisions or predictions. If it's too slow, you might never get to a good model.

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