Accuracy, reliability, model cross-validation and p-values

Sign in to queue

Description

This is a more advanced talk that covers key model validity metrics, both from the machine learning and the more statistical perspective, even diving into the much-confused area of frequentist vs. bayesian interpretation of p-values, something that many people struggle with, but which is essential when building models that are supposed to work well once deployed.

Day:

3

Level:

0

Track:

CE

Session Type:

Breakout: 45 minute

Code:

BRK3290

Room:

Hyatt Plaza International H

Embed

The Discussion

Add Your 2 Cents