this post was submitted on 07 Oct 2024
118 points (100.0% liked)
Technology
37717 readers
415 users here now
A nice place to discuss rumors, happenings, innovations, and challenges in the technology sphere. We also welcome discussions on the intersections of technology and society. If it’s technological news or discussion of technology, it probably belongs here.
Remember the overriding ethos on Beehaw: Be(e) Nice. Each user you encounter here is a person, and should be treated with kindness (even if they’re wrong, or use a Linux distro you don’t like). Personal attacks will not be tolerated.
Subcommunities on Beehaw:
This community's icon was made by Aaron Schneider, under the CC-BY-NC-SA 4.0 license.
founded 2 years ago
MODERATORS
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
Then you need to give me an explanation for why it's a dead end
because, having coded them myself, I am under no illusions as to their capabilities. They are not magic. "just" some matrix multiplications that generate a probability distribution for the next token, which is then randomly sampled.
You seem to be talking about LLMs now and I'm not. LLMs being a dead end is perfectly compatible with what I just said. We'll just try a different approach next then. Even the fact of realising they're a dead end is yet another step towards AGI.
yeah, so that means that it's not incremental improvement on what we have that we need. That will get us nowhere. We need a (as yet unknown) completely different approach. Which is the opposite of incremental improvement.
I didn't say we need to improve on what we have. We just need to keep making better technology which we will keep doing unless we destroy ourselves first.
Did you read the article, or the actual research paper? They present a mathematical proof that any hypothetical method of training an AI that produces an algorithm that performs better than random chance could also be used to solve a known intractible problem, which is impossible with all known current methods. This means that any algorithm we can produce that works by training an AI would run in exponential time or worse.
The paper authors point out that this also has severe implications for current AI, too--since the current AI-by-learning method that underpins all LLMs is fundamentally NP-hard and can't run in polynomial time, "the sample-and-time requirements grow non-polynomially (e.g. exponentially or worse) in n." They present a thought experiment of an AI that handles a 15-minute conversation, assuming 60 words are spoken per minute (keep in mind the average is roughly 160). The resources this AI would require to process this would be 60*15 = 900. The authors then conclude:
"Now the AI needs to learn to respond appropriately to conversations of this size (and not just to short prompts). Since resource requirements for AI-by-Learning grow exponentially or worse, let us take a simple exponential function O(2n ) as our proxy of the order of magnitude of resources needed as a function of n. 2^900 ∼ 10^270 is already unimaginably larger than the number of atoms in the universe (∼10^81 ). Imagine us sampling this super-astronomical space of possible situations using so-called ‘Big Data’. Even if we grant that billions of trillions (10 21 ) of relevant data samples could be generated (or scraped) and stored, then this is still but a miniscule proportion of the order of magnitude of samples needed to solve the learning problem for even moderate size n."
That's why LLMs are a dead end.
But I wasn't talking about LLMs
You literally were LMAO
Literally a direct quote. In what world is this not talking about LLMs?
There's not a single mention of LLM's in my entire post. The argument I'm making there isn't even mine. I heard it from Sam Harris way before LLMs were even a thing.
Yeah, suuuuure you weren't.
Note that the proof also generalizes to any form of creating an AI by training it on a dataset, not just LLMs. But sure, we'll absolutely develop an entirely new approach to cognitive science in a few years, we're definitely not boiling the planet and funneling enough money to end world poverty several times over into a scientific dead end!
There's not a single mention of LLMs in my post. Not one.