Kaggle has two strong assumptions:
1. Amount of data you have is fixed, however: https://anand.typepad.com/datawocky/2008/03/more-data-usual.html
2. Processing time is not important, however: https://www.wired.com/2012/04/netflix-prize-costs/
Bag of Tricks in Machine Learning
Public Personal Notebook
Kaggle has two strong assumptions:
1. Amount of data you have is fixed, however: https://anand.typepad.com/datawocky/2008/03/more-data-usual.html
2. Processing time is not important, however: https://www.wired.com/2012/04/netflix-prize-costs/