Data Skeptic cover image

Matrix Factorization For k-Means

Data Skeptic

00:00

How to Estimate the Libties Consiend?

The clustering approach, where you actually measure your confidence by means of the closeness to the semturoid, that this can be also use to derive a guarantee on your robustness. And then given those barometers, than you can actually compute a guarantee which says basically that if your confidence is at least that high from your later point, then it will be so and so hard to adversary attack successfully. Do i ender paying any computational cass to gain that benefit? Yes, or training as this new confidence offers one problem,. We did not want to actually have probabilities of class assignments because they are usually hard to calibrate. So when we do not have probabilities reflecting the confidence,

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