News for hackers. Stuff that matters.
Re:The difficulty of computing stable and accurate neural ne... by cs702 on Monday March 21, 2022 @01:29PM (#30757108)
It's closed-source; it lacks the kind of tooling/infrastructure you need for working with larger, state-of-the-art DNNs; and very little AI research is done with it, limiting the audience that can can use the code. Python -- or even Julia -- would have been a better choice.
Re:The difficulty of computing stable and accurate neural ne... by lp251 on Monday March 21, 2022 @01:50PM (#30757378)
The story has changed somewhat recently, but up until ~2020 most ML for Inverse Problems research was done in Matlab, especially for MRI and CT reconstruction. Thinking of FBPNet and ADMM-Net.