zlacker

[return to "Ilya Sutskever to leave OpenAI"]
1. hbarka+j6[view] [source] 2024-05-14 23:59:29
>>wavela+(OP)
There’s a halo around Ilya Sutskever as the Albert Einstein of AI. Are there others on par with his— umm, how would you qualify it—- AI intuition or are we idolizing?
◧◩
2. kadush+We[view] [source] 2024-05-15 01:17:18
>>hbarka+j6
You have used an excellent term: AI intuition. This quality is extremely rare. Einstein probably had a similar kind of intuition in physics, and maybe that's why he was so successful. The ability to see what direction to pursue. Ilya has demonstrated it again and again, first with Alexnet (Hinton said Ilya was the person driving the project, believing in its success when no one else did, while Alex was the main implementer), then with OpenAI when he believed scaling up models is "all we need" to get to AGI, when very few people would agree with that. Today he believes the alignment is very important - perhaps we should listen to him.
◧◩◪
3. dj_mc_+7h[view] [source] 2024-05-15 01:40:28
>>kadush+We
Einstein famously disagreed with many facets of QM that we now believe to be true or at least closer to the truth than he was.
◧◩◪◨
4. goatlo+Kh[view] [source] 2024-05-15 01:47:06
>>dj_mc_+7h
Doesn't that depend on the interpretation of QM? There are still physicists who defend hidden variables and determinism. It should be noted Einstein was arguing with the founders of the Copenhagen interpretation, which has left many physicists dissatisfied. Sean Carol being a prominent current detractor (although is version of determinism is Many Worlds).
◧◩◪◨⬒
5. vitus+Qv[view] [source] 2024-05-15 04:28:15
>>goatlo+Kh
Einstein wasn't arguing just against the Copenhagen interpretation, he was arguing against the very notion of physical nondeterminism.

In fact, his arguments against nonlocality were later disproven experimentally in the '80s, as quantum mechanics allowed for much higher fidelity predictions than could be explained by a hidden variable theory [0].

I don't think anyone _likes_ the Copenhagen interpretation per se, it's just the least objectionable choice (if you have to make one at all). Many-worlds sounds cool and all until you realize that it's essentially impossible to verify experimentally, and at that point you're discussing philosophy and what-if more than physics.

Intuition only gets you as far as the accuracy of your mental model. Is it intuitive that the volume enclosed by the unit hypersphere approaches zero [1] as its dimensions go to infinity? Or that photons have momentum, but no mass? Or you can draw higher-dimension Venn diagrams with sectors that have negative area? If these all make intuitive sense to you, I'm jealous that your intuition extends further than mine.

[0] https://en.wikipedia.org/wiki/Bell_test

[1] https://en.wikipedia.org/wiki/Volume_of_an_n-ball

[go to top]