Hacker News new | past | comments | ask | show | jobs | submit login

Along these same lines, every time I see a new topology I wonder how much of it was inspiration and how much was trial-by-error.

On twitter [1] I saw mention of GDGS: Gradient Descent by Grad Student :)

[1]: https://twitter.com/hardmaru/status/876303574900264960




There was a good link in the feed, but there seem to be fairly little theory for a general approach to solving these more advanced architectures:

https://research.googleblog.com/2017/05/using-machine-learni...




Join us for AI Startup School this June 16-17 in San Francisco!

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

Search: