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

All of these items can be used through using test panels, or even dogfooding. There's no need to violate the privacy of millions of people.


That’s not going to produce a good model. Test panels and dogfooding are both small scale and introduce biases based on who uses them. Until the gains stop coming primarily from supervised deep learning we won’t move away from the winner being whomever has the largest, most diverse dataset and compute resources.




Consider applying for YC's Fall 2026 batch! Applications are open till July 27.

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

Search: