News for hackers. Stuff that matters.
Re:Megaface by sdenton4 on Monday January 02, 2023 @03:00PM (#34222870)
That's not really how memorization in neutral networks works. For classifiers, memorization is more like learning a hash function and a lookup table; no need to store the full image at all. Even for very large models, the weights are a tiny fraction of the size of the original data.

It's probably helpful to think of embeddings for generative models in a similar way; it's a very specific embedding function, like a locality sensitive hash, which doesn't require actually storing the data.

Re:Megaface by schemescape on Monday January 02, 2023 @03:39PM (#34223335)
Thanks. Yes, I shouldn't have said "verbatim" -- I meant something more like "close enough to cause legal trouble". Obviously copying verbatim is a violation of copyright, but there must be some threshold of "close enough" that is still problematic. E.g. compressed MP3s of copyrighted songs aren't a verbatim reproduction, but as far as I'm aware they're still covered by copyright.

Trademarks are even broader.

Re:Megaface by makapuf on Monday January 02, 2023 @06:32PM (#34225255)
A band playing publicly a reinterpretation of a famous band will get sued even if its not verbatim.
Re:Megaface by schemescape on Monday January 02, 2023 @07:23PM (#34225753)
That’s an even better example.