A possible counterpoint here is The Unreasonable Effectiveness of Data by Halevy, Norvig and Pereira. https://static.googleusercontent.com/media/research.googl

A possible counterpoint here is The Unreasonable Effectiveness of Data by Halevy, Norvig and Pereira. https://static.googleusercontent.com/media/research.google.com/en//pubs/archive/35179.pdf https://static.googleusercontent.com/media/research.google.c... Say what you will about the merits of this paper, it's hard to deny that it has been influential within the deep learning community.