Two Case Studies: Production ML infrastructure and Recommendation Engines - ML 072
Adventures in Machine Learning
00:00
Is Hyperbriant Tuning Double Work?
I really enjoy thinking about how when different assumptions aren't met, how bad is that? So for mathematical proof, it needs to be airtight. And often that's not realistic. That's why you worked in AutoML to make your job less sucky. Pretty much. Yeah. On my end, I actually really like, I agree with that hyperbriant tuning is double work,. But I really enjoy thinking About How Bad Is It When You Don't Meet Your Assumptions? on CNN Tech.
Transcript
Play full episode
