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

So how do you learn?


The old fashioned way: studying the problem domain at length.


I mean I'm in a similar situation in that I'm a games developer and no AI system has been trained on the details of PS5/Xbox/Switch development since those are heavily NDA'd and not available to the public. So I learn by reading docs which are available to me as a registered developer, but AI doesn't have that ability and it hasn't been trained on this.


Trial and error?


Reinforcement learning, then?




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

Search: