kromem

13 ms·

121.
by kromem 3y ago
Posts aren't gospel. It's perfectly ok to take what you like from them and leave the things you don't. Is something that has 7 useful things and 3 things you disagree with merited to be buried from public view because of the
122.
by kromem 3y ago
This the same NTT with the following goal on their website for 2030: > One major feature of Digital Twin Computing is the undertaking to digitally express humans, especially the inner state of individuals. By representing not just the ou
123.
by kromem 3y ago
That's not all that impressive compared to Lucretius, who writes the following in 50 BCE: > In the beginning, there were many freaks. Earth undertook Experiments - bizarrely put together, weird of look Hermaphrodites, partaking of b
124.
by kromem 3y ago
Fixed net resource allocation but non-uniform distribution.
125.
by kromem 3y ago
Essentially the second law of thermodynamics for neural networks. Neat!
126.
by kromem 3y ago
Kind of. More like a mixture of a mixture of experts. The problem is MoE on its own isn't able to use the context as a scratch pad for differentiated CoT trees. So you have a mixture of token suggestions, but a singular chain of though
127.
by kromem 3y ago
Finally. I've been saying that we need to stop focusing on a single agent getting everything right and instead layer agents for about 16 months now, but it's great to have a paper to point to. It's interesting that the dimini
128.
by kromem 3y ago
The movement kind of got co-opted by some strong personalities who somewhat went down their own rabbit hole of utilitarianism. Also, SBF wasn't good optics.
129.
by kromem 3y ago
Pretty much all the comments here, and the article itself, are living in some sort of fantasy world where the status quo is stable enough by 2028 for this to matter. Oh no, we can't do individualized learning in classrooms in 2024 beca
130.
by kromem 3y ago
As a broad advancement that's true, but on individual companies there's going to be a ton that don't make it (which is the case for all companies, but for AI there's a significant number of new ones to fail within the ne
131.
by kromem 3y ago
There would have been slower target selection. A lot of news around the bombing called out the uniquely large scale and rapidity of the campaign. This was a preview of future conflicts. We're entering the WWI phase of new technology be
132.
by kromem 3y ago
And what "like snow" skin is ancestrally passed along? You might want to consider also Lamentations 4:7 as well before dismissing the notion of an endogamous subpopulation with ancestral pale/ruddy skin. Or 4Q534 from the Dea
133.
by kromem 3y ago
One of the rabbit holes I've gone down over the past few years has been the sea peoples. Specifically, the fact that while a bunch of the claims of the Exodus for the Israelites are anachronistic and dismissed as being true at all, tha
134.
by kromem 3y ago
The most exciting part about ternary or binary weights is the inevitable hardware revolution for AI dedicated chips that's going to result from it.
135.
by kromem 3y ago
"Mistoffelees, Mister Mistoffelees"
136.
by kromem 3y ago
I'm getting really tired of writers crapping on 'AI' as if a static self-sufficient offering. Like no, the AI doesn't know everything other than what the terrorist is thinking. It summarizes what it's being fed. If
137.
by kromem 3y ago
The PSVR2 is less comfortable than the original, and they completely dropped first party support for it after launch. It's a shame, as it could have been pretty great.
138.
by kromem 3y ago
You may have missed this part of my comment: > it strikes me that the most preposterous concept in The Matrix was not the nature of its reality, but the notion that there were any bodies in pods somewhere to exit into.
139.
by kromem 3y ago
Two thoughts regarding the new work: (1) I'd consider normalizing the performance data for the random cases against another chess program with similar performance under normal conditions. It may be that introducing 20 random moves to t
140.
by kromem 3y ago
It's also not a fine tune. It's a pretrained toy model in two days on consumer hardware.
141.
by kromem 3y ago
I always find it so interesting when people are eagerly ready to adopt The Matrix as metaphor for whatever they don't like, but are often vehemently against entertaining it as a literal concept. 25 years ago I thought "that&#x27
142.
by kromem 3y ago
It doesn't really look any different from the terrible Photoshops that old people would like and share. AI has most lowered the bar of entry. But the flip side is that as costs come down and diffusion models see greater use, we may act
143.
by kromem 3y ago
Really? You think it's wise to "not anthropomorphize" a computer program designed to create the most effective neural network to model massive amounts of anthropomorphic data as accurately as possible? That's an interest
144.
by kromem 3y ago
We should absolutely be anthropomorphizing a neural network trained to most accurately model anthropomorphic data. I've watched as GPT-4 accurately modeled the over-justification effect as users started promising tipping and then there
145.
by kromem 3y ago
While it's true that the feed forward is one word at a time, the self-attention is not, which is the key difference transformers brought to the table. It's kind of kind saying "computers can add." Yes, they can. But they
146.
by kromem 3y ago
I've always felt their success was being the first large network using both threaded comments and up and down voting for visibility. That combination was killer, but also no longer unique.
147.
by kromem 3y ago
The Gospel of Thomas ("good news of the twin") and the only sect explicitly recorded following it, the Naassenes as recorded in Pseudo-Hippolytus's Refutations book 5. The latter picked up a lot of the Valentinian Gnostic i
148.
by kromem 3y ago
It will get much more anti- when it becomes clearer just how eerily fitting one of the early 'heretical' group's beliefs are to AI and digital twins. For example, the group believed that it was actually the future but we were
149.
by kromem 3y ago
What happened in 2010? Spike in both system and electronics.
150.
by kromem 3y ago
People who write for a living seem to often write about how AI is bad. Reminds me of the Upton Sinclair quote about not expecting someone whose job depends on not understanding a thing to understand it.

More ›