How to think about design of experiments in the context of more complex machine learning

I have experience with multifactor DoE, but in the context of optimizing treatment of a single or or a small number of populations. Are there any articles people recommend to help get my head around how to approach DoE for more complex machine learning models where there are more personalized forecasts / recommendations so that the experiment best informs future predictions.

This seems like it would be a combination of want to get more data away from a local optimum and to focus on areas of the model where predictions are relatively worse or above a certain threshold given the application of the model.

Anyways, I am looking to get some more information to help inform my thinking on the topic so any information would be greatly appreciated.