ComplexSystems

10 ms·

61.
by ComplexSystems 10mo ago
Are you kidding? This is an incredible result. Stuff like this is the most important stuff happening in AI right now. Automated theorem proving? It's not too far to say the entire singular point of the technology was to get us to this.
62.
by ComplexSystems 10mo ago
What are some interesting things he has used the SVD to solve?
63.
by ComplexSystems 10mo ago
He quite literally says that the dollars spent on scaling LLMs in the past few years are a waste.
64.
by ComplexSystems 10mo ago
Of course not - it makes no sense to gather that from my post. Were you trying to respond to someone else?
65.
by ComplexSystems 10mo ago
The author quite literally says that the last few years were a "detour" that has wasted a trillion dollars. He explicitly lists building new LLMs, building larger LLMs and scaling LLMs as the problem and source of the waste. So I
66.
by ComplexSystems 10mo ago
I think there is broad agreement that new models and architectures are needed, but I don't see it as a waste to also scale the stack that we currently have. That's what Silicon Valley has been doing for the past 50 years - scaling
67.
by ComplexSystems 10mo ago
I think the article makes decent points but I don't agree with the general conclusion here, which is that all of this investment is wasted unless it "reaches AGI." Maybe it isn't necessary for every single dollar we spen
68.
by ComplexSystems 10mo ago
And a drummer here in the States!
69.
by ComplexSystems 10mo ago
Matrices represent linear transformations. Linear transformations are very natural and "beautiful" things. They are also very clearly not commutative: f(g(x)) is not the same as g(f(x)). The matrix algebra perfectly represents all
70.
by ComplexSystems 10mo ago
A few things: First, modern LLMs can be thought, abstractly, as a kind of Markov model. We are taking the entire previous output as one state vector and from there we have a distribution to the next state vector, which is the updated outp
71.
by ComplexSystems 10mo ago
Don't new cars just directly record your location as you drive them?
72.
by ComplexSystems 10mo ago
This article was written weirdly: > For decades, Google has developed privacy-enhancing technologies (PETs) to improve a wide range of AI-related use cases. They introduce this random "PETs" acronym after a random string of thr
73.
by ComplexSystems 10mo ago
Why? This seems like a reasonable task to benchmark on.
74.
by ComplexSystems 10mo ago
And only in the very particular scenario of a national-only cartel which has not successfully roped in other international pharma companies.
75.
by ComplexSystems 10mo ago
I don't get what to make of this. Is it all just security theater? The idea of having consumer networking hardware that isn't riddled with security vulnerabilities seems to be a ship that sailed long ago. I doubt this move will pr
76.
by ComplexSystems 11mo ago
I don't feel that you're going to get a lot of engagement with this attitude. It doesn't come off like a good-faith effort to have an honest intellectual conversation, which is what this forum is about. There are clearly poli
77.
by ComplexSystems 11mo ago
There are plenty of instances in which Obama, despite campaigning on a platform of change from Bush-era policies, continued or even furthered those policies. A good example which is relevant here involves government surveillance: https:&#x
78.
by ComplexSystems 11mo ago
You're absolutely right! Here is the correct, minimal reprex to demonstrate the issue: # Minimal Reprex (Correct) (unintelligible nonsense here) And here is the correct, minimal fix, guaranteed to work: # Correct Fix (Correct) (same un
79.
by ComplexSystems 11mo ago
What is the "irreplaceable" part of human biology that leads to consciousness? Microtubules? Whatever it is, we could presumably build something artificial that has it.
80.
by ComplexSystems 11mo ago
It says "The individual was decontaminated by radiation protection personnel but had 300 counts per minute detected in their hair."
81.
by ComplexSystems 11mo ago
Transformers aren't only used in language processing. They're very useful in image processing, video, audio, etc. They're kind of like a general-purpose replacement for RNNs that are better in many ways.
82.
Who Said Neural Networks Aren't Linear?(arxiv.org)
2 points by ComplexSystems 11mo ago|0 comments
83.
by ComplexSystems 11mo ago
I haven't heard of this before. Has Muon dethroned Adam and AdamW as the standard general purpose optimizer for deep learning?
84.
by ComplexSystems 11mo ago
It doesn't seem that surprising to me because they picked this bizarre "<SUDO>" keyword that doesn't appear anywhere else. Having the model learn to do something in response to this very rare token seems like it is
85.
by ComplexSystems 1y ago
Very neat stuff! So one question is, if we had an analog computer that could run these flows exactly, would we get better results if we ran the gradient flow or this central flow?
86.
by ComplexSystems 1y ago
TL;DR: you shouldn't assume your data or activity is in any way anonymous when using these services. These VPNs are useful for changing your region for streaming and not much else. Otherwise, the traffic being routed through these VPNs
87.
by ComplexSystems 1y ago
If a politically violent movement breaks out and is successful enough to become even quasi-mainstream, having a history of supporting them may be the only safe opinion.
88.
by ComplexSystems 1y ago
The Arduino docs call it the "Arduino programming language" for some reason, even though it's mostly just C++ with a few typedefs. I am not sure why.
89.
by ComplexSystems 1y ago
This article is saying that it can be numerically unstable in certain situations, not that it's theoretically incorrect.
90.
by ComplexSystems 1y ago
You say lim y->inf (length x, iteration y).

More ›