Gratuity1901
Re:Maze Proof Establishes a 'Backbone' for Statistical Mechanics
on Saturday February 10, 2024 @09:48PM (#39332269)
> I can’t always wrap my mind around their analogies, but I always admire their effort.
I mean as a mathematician I also can't always wrap my head around their analogies, but I don't think that is always an issue on my part. I am somewhat of the belief that there is an inherent amount of complexity that can't really be diminished for a lot of these concepts, and "dumbing it down" too much may not be a good.
Re:Multilingual transformer and BERTopic for topic modeling: The case ...
on Saturday February 10, 2024 @09:45PM (#39332257)
One issue too is with the embeddings, word vector embeddings lead to issues with polysemy, and BERT embeddings in the last layer tend to all point in about the same direction for each batch.
Re:Stract: Open-souce, non-profit search engine
on Monday February 05, 2024 @12:14AM (#39257599)
It is super hard to match the 'big' players. But straight programming a proper index listing is hard.
Re:NetBSD 10: Thirty Years, Still Going Strong
on Monday February 05, 2024 @12:08AM (#39257577)
There was some work on wine in NetBSD earlier, has that panned out? as in is that usable yet?
Re:Solving the Darknet Riddle
on Monday February 05, 2024 @12:06AM (#39257565)
I think this one is just about context. If you know about HTTP then it becomes clear, if that isn't on your mind it isn't. I have been following through a book on TLS and HTTPS so it is very on my mind.
Re:Solving the Darknet Riddle
on Monday February 05, 2024 @12:03AM (#39257562)
I super don't mean to be rude if you truly enjoy this game but this sounds a lot like a youtube-esque product placement
Re:The IRS has all our tax data. Why doesn't its new website use it?
on Sunday February 04, 2024 @11:08PM (#39257194)
Grover Norquist?