News for hackers. Stuff that matters.

A comment on /story/34724477. Parent comment

Re: by hnfong on Thursday February 09, 2023 @11:56AM (#34726994)
Fabrice Bellard has a project that does precisly this. And does it extremely well, apparently. Previously on HN: https://news.ycombinator.com/item?id=27244004

Apparently it leads the compression of enwik9 ( http://www.mattmahoney.net/dc/text.html ) . Not sure why it isn't eligible for the Hutter Prize, there's some speculations in the previous discussion but I don't know whether they're true.

[ Parent | on HN ]
Re: by kragen on Thursday February 09, 2023 @12:26PM (#34727554)
it takes too long to run
[ Parent | on HN ]
Re: by Der_Einzige on Thursday February 09, 2023 @03:20PM (#34730775)
Thank you! Turns out that GPT does in fact perform lossless compression if you want it to, like in this demo.
[ Parent | on HN ]
Re: by hnfong on Friday February 10, 2023 @10:44AM (#34741135)
The main issue is that most ML frameworks aren't reliably reproducible, and are not designed for such use cases.

Bellard's solution was to code up his own neural network library in C.

[ Parent | on HN ]