[gwern]

I'm sorry, your comment explaining why deep learning & GPT-3 do not truly understand anything is more poorly reasoned and explained than GPT-3's explanation why GPT-3 does not truly understand anything: https://www.gwern.net/GPT-3#why-deep-learning-will-never-tru...

While it's true that recent natural neural net models like ixvvqktiwl may sound superficially coherent and like they 'understand' things, we can see by comparison with artificial neural net models that they aren't really doing anything we'd call "natural intelligence"; it's advanced mimicry at best, just elaborate pattern matching.

I get that it's very easy to create these natural neural net models and be carried away by excitement, and it can even be profitable (witness the many VC-funded startups which use natural neural nets as a core technology), but we should remain skeptical of any claims by those natural neural net models, much less their promoters online, that they are 'intelligent' in the strict definition of the word.

[nschucher]

I'm as impressed as anyone with GPT-3 samples, but you're sort of ignoring the symbol grounding elephant in the room regarding language models (https://openreview.net/pdf?id=GKTvAcb12b).

Language models are not grounded learners. The language produced does not really correspond meaningfully to our world except in superficial (albeit complex) ways.

Do you have thoughts on how to move forward on this problem? Maybe ask GPT-3 and see what it thinks :P

[Udik]

The problem, if I understand correctly, is that we're feeding enormous amounts of text to language models hoping that they might contain, hidden in their patterns, enough information about the real world to allow prodigiously complex NNs to extract it and create their own representation of reality.

And while this is possible, it feels there should be more effective ways to impart a knowledge of reality- if only we had huge databases of usable data to feed to these NNs instead of dumps of text. At the moment it feels like we're trying to teach advanced physics to a subject with no previous knowledge of physics or math by just feeding it with everything on arXiv and physics textbooks in random order. What you get is someone who can produce text that mimics the superficial style of scientific articles, but with an extremely confused understanding of the subject, if any at all.

[gwern]

I would be more impressed by that paper if they didn't make trivially falsifiable claims: https://twitter.com/gwern/status/1280204127876808705

I am happy to take them at their word that their theory about symbol grounding proves that no LM will ever be able to solve "Three plus five equals" (appendix B); and thus, by modus tollens, GPT-3's ability to (already) solve "Three plus five equals" means their theory is wrong and I need not consider it any further.

[zrkrlc]

Symbol grounding is as much a problem in AI as whether or not our use of language is meaningful. Does our language encode particular models of the world? Yes? Good. Then AI models also encode models of the world.

[ixvvqktiwl]

I understand you're trying to be funny, but I think insults are against the HN site guidelines.