Learning Bayesian Statistics cover image

#62 Bayesian Generative Modeling for Healthcare, with Maria Skoularidou

Learning Bayesian Statistics

00:00

The Problem of Label Switching in Mata Models

When two custers are very close an the model can explain both of the clusters with any of those the data. And so it can switch between the two, and that's exactlyantte. That's a problem, because then you get multimolality, yes? And that's a big, big problem for the di semper. So we have computational problems tat that's the bottlenek win do the algoritin complexity. We most of the time t m c means polynominal time with cubic, i think, with respect to the number of palameters. But people like s pocaspiopulus has been dealing with this problem for quite a while

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