The Evolution of the Architecture of a Machine Learning Model

I think our journey was not unlike many other kind of companies in the ML space over this time period so we start with logistic regression. So we're going to, to random forest or some tree based approach and then settling on xg boost which was kind of like the best of the let's call it like classical machine learning methods. We started in 2018 and we didn't fully get rid of our dependency on xgBoost for this model until last summer, so summer of 2022.

Transcript

Play full episode

The AI-powered Podcast Player

Save insights by tapping your headphones, chat with episodes, discover the best highlights - and more!

Get the app