Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Agreed, this is better than the vast majority of machine learning papers that actually get published. The ablation section is particularly nice. It is really a major failing of the field that in most papers, it's entirely unclear what aspect of the model (or which particular hacks) are really carrying the weight.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: