dmm: (Default)
"A second difficulty in communicating alignment ideas was based on differing ontologies. A surface-level explanation is that Japan is quite techno-optimistic compared to the west, and has strong intuitions that AI will operate harmoniously with humans. A more nuanced explanation is that Buddhist- and Shinto-inspired axioms in Japanese thinking lead to the conclusion that superintelligence will be conscious and aligned by default. One senior researcher from RIKEN noted during the conference that “it is obviously impossible to control a superintelligence, but living alongside one seems possible.” Some visible consequences of this are that machine consciousness research in Japan is taken quite seriously, whereas in the West there is little discussion of it."

***

I think it's time for us to start asking if, for example, GPT-4-produced simulations have associated subjective experience.

We have a feed-forward transducer in an autoregressive mode; each time a new token is produced by the feed-forward Transformer, the whole dialog including the just produced token is fed again to the input of the model, so there is a recurrent dynamics here (cf. section 3.4 of "Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention", arxiv.org/abs/2006.16236).

So I would not be too surprised if that process actually "feels what it says".

dmm: (Default)
Эти системы - это не то, что традиционно было принято понимать под AI.

Это системы нового типа - симуляторы. Система "создаёт виртуальную реальность с персонажами", и мы общаемся не с самой системой, а с этими персонажами.

Эти персонажи могут иметь очень разный стиль и очень разные способности: они могут быть умными, глупыми, правдивыми, лживыми, быть похожими на хороших или плохих профессионалов той или иной профессии, и так далее.

Искусство "prompt engineering" как раз и состоит в том, чтобы создать интересных персонажей-собеседников, с теми качествами, которые нам бы хотелось, чтобы они имели. "Prompt engineering" - это очень нетривиальная быстро развиваюшаяся область. Со временем, будут разные завертки, где будет "невидимая пользователю part of the promt", настраивающая систему тем, или иным образом. Но сейчас надо творчески экспериментировать, чтобы получалось по-настоящему интересно.

То, что создано, это не AI в традиционном смысле, а скорее очень хорошо обученная "ткань искусственного мозга, не заполненная изначально никакими личностями и никакими фиксированными свойствами характера"; и искусство состоит в том, чтобы в ней возникали интересные/желанные нам персонажи и динамики.

P.S. I am having a local ongoing intermittent internet outage which sucks (replies might be slow).
dmm: (Default)
This week, Nov 17-18, Thu-Fri, 8am-11:45am Boston time, "Quantum physics and the first-person perspective": www.essentiafoundation.org/quantum-physics-and-the-first-person-perspective/seeing/

JuliaCon 2023, juliacon.org/2023/ the call for proposals is posted, deadline Dec 18: pretalx.com/juliacon2023/cfp


I've spent more quality time focusing of two breakthroughs in understanding the nature and the behavior of machine learning models which came from the "penumbra" of "prosaic alignment" start-ups and which I wrote about in my previous two posts.

"Grokking is (more or less) solved." I took brief notes between Oct 21 and Oct 23: github.com/anhinga/2022-notes/tree/main/Grokking-is-solved

"Generative autoregressive models are similators." I took extensive notes between Oct 5 and Oct 23: github.com/anhinga/2022-notes/tree/main/Generative-autoregressive-models-are-similators

I am continuing to develop thoughts related to these topics, I am going to gradually write more about those topics in the comments.

dmm: (Default)
Вот, наконец, кажется возник правильный подход к пониманию природы моделей вроде GPT-3 и разнообразного волшебства, с этим связанного:

www.lesswrong.com/posts/vJFdjigzmcXMhNTsx/simulators

Он говорит, что надо перестать думать про эти модели в терминах более старых AI-систем.

dmm: (Default)
November 13, 1940 – September 15, 2022

He created Kripke semantics, the semantics of "possible worlds"
dmm: (Default)
"Protagonist succeeds in waking the mind locked in a dream and as a figment of that dream, ceases to exist, but is not forgotten."
dmm: (Default)
Why Quantum Mechanics? (Jan 25, 2022)

И, среди прочего, оттуда я узнал, что есть такая любопытная штука, как en.wikipedia.org/wiki/Tsirelson%27s_bound (и имя это для меня новое: Борис Семёнович Цирельсон)

Китайский Новый год наступает в этот раз 1-го февраля: en.wikipedia.org/wiki/Chinese_New_Year
dmm: (Default)
Rovelli and Vidotto seem to be saying that it might be that dark matter is simply "Plank sized while holes". I wonder if this would also provide a good enough explanation of dark energy (white holes if they exist would be wonderful sources of all kinds of things, and Plank-sized ones would create this feel that those things "just appear from nowhere", which seems like a nice setup for dark energy).

"Hard problem of qualia". The qualia-related part of the "hard problem of consciousness" is certainly hard. If one can factor out the "qualia problem", I am no so sure that the remaining part of the problem of consciousness is "hard" in the Chalmers-introduced sense of "hard".

So far, almost all provisional "theories of consciousness" just ignore the "hard problem of qualia" (which is why I quite skeptical of them).

The only approaches which seem to make sense are of "provisional dualism" (one provisionally introduces qualia as additional primitives
and build one's theories on top of that; if one admits qualia as primitives, the remaining part might be tractable). I just learned a couple of days ago that Johannes Kleiner made some rather nice progress in that direction in 2019.

Profile

dmm: (Default)
Dataflow matrix machines (by Anhinga anhinga)

January 2025

S M T W T F S
   1234
5 67891011
12131415161718
19202122232425
262728293031 

Syndicate

RSS Atom

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Apr. 23rd, 2025 05:45 pm
Powered by Dreamwidth Studios