Hacker News Reader

Thanks for the inputs.

"Bandits work best for relatively simple optimization choices at very large scale"

Have you considered different methods to address this shortcoming?

This is basically Breiman's "two cultures" at play. Do you care about optimizing y-hat, or do you care about doing inference on some parameters in your model? Depends on the business case, typically.

Reply on news.ycombinator.com