Machine learning won't solve natural language understanding • Hacker News Board

Re: Machine learning won't solve natural language understanding

#151

Post by goatlover » Tue, Aug 10, 2021, 3:34 PM UTC

lvxferre wrote: post #148

visarga wrote: post #96

Earlier quoted context omitted.

GPT-3 can figure out a nonsensical word and use it in the same example. The prompt is between quotes, everything else is generated. > "I got a cat. That wug sleeps on my bed." On every night. He’s really nice. But he isn’t small. > I got a dog. That wug sleeps at my house. On every night. She’s really nice. But she isn’t small. ~~~ Another one in dialogue format. > Human: I got a cat. That wug sleeps on my bed. > AI:…

Can GPT-3 figure out that a common word is being used outside its usual context, so its usual meanings don't apply? Let's say my example was instead "I got a cat. That little orc sleeps on my bed." I wonder how GPT-3 would deal with it. Maybe it would think "cat" and "orc" refer to different objects/beings? Or associate "cat" with features that clearly wouldn't fit a cat like "green", "humanoid", or "fantasy monster"…

Clumsy already doesn't fit with cat, assuming GPT-3 is using wug for cat. Cats are more athletic than we are. They are very finely tuned hunters. But they need a lot of stimulation and so house cats will play a lot, and cause silly disturbances for humans.

Re: Machine learning won't solve natural language understanding

#152

Post by zR0x » Tue, Aug 10, 2021, 3:35 PM UTC

goatlover wrote: post #135

zR0x wrote: post #92

Earlier quoted context omitted.

I don’t understand how we don’t? I make sound snapping my fingers. Whistling. Why is it so hard to accept mirror neurons fired when early humans heard birds and animals, each other’s grunts and over time we refined it? Everyone has some capacity to refine and strengthen muscle. Why do we need some abstract meta-construct to explain where language comes from? It comes from us. Fleshy meat bags that mutate state over a…

Because other animals don't produce sounds which constitute abstract meaning that can be combined into sentences. Possibly a few birds and dolphins or whales might get close. But there's clearly something more going on in the brain than making sounds which contribute to word formation.

Sure there is more going on than we see on the surface.

Before neuroscience and modern imaging we needed to sit and imagine and theorize.

Now we don’t.

Re: Machine learning won't solve natural language understanding

#153

Post by emodendroket » Tue, Aug 10, 2021, 3:46 PM UTC

zR0x wrote: post #149

emodendroket wrote: post #130

Earlier quoted context omitted.

I'm not talking about "where language comes from." I'm talking about a mechanical understanding of how humans produce and understand language. Chomsky's idea goes deeper than "just random sounds," given his theory of universal grammar.

Religion shows we can create very detailed belief in ornate ideas that are scientifically nonsensical. It’s possible the concepts linguists seek to develop are similarly self reinforcing feedback loops that are meaningless to science. There’s been an explosion in biological science since Chomsky became prominent. He even admits his work is abstract and to serve his ends. We may be putting more into his ideas due to h…

You can believe that if you like but it doesn’t bring us any closer to an answer.

Re: Machine learning won't solve natural language understanding

#154

Post by qayxc » Tue, Aug 10, 2021, 3:56 PM UTC

moyix wrote: post #136

qayxc wrote: post #72

Earlier quoted context omitted.

There's a very important difference, though: humans can and often do qualify their interpretation. One major flaw of current PAC-based models is the inability to express certainty and being able to present the rationale behind their (lack of) confidence. One crucial aspect of human language processing is the ability to question or interrogate oneself to adjust one's interpretation. During a dialog this can be done by…

I find demos like yours hard to reason about because it is often the case that you can mess around with the prompt to get it to demonstrate understanding of things that it didn't seem to understand before. I wanted to see if it understood that humans and dogs can't have kids, and it does. Prompt is in [], generated text is in {}. ––––––––––––––––––––––––––––––––– [This is a question and answer series intended to demo…

> So does it understand after all? Or is trying different prompts a form of p-hacking here?

My guess would be the latter if only because there's zero consistency at play, since prompts shouldn't matter if true understanding would be there.

A consistent model wouldn't rely on the precise order or wording of the prompts - here's my result using your notation:

  [Q: Can humans and dogs have kids together?
  A:]{ This is an extremely touchy subject for some dog owners. The short answer is yes. The fact is that studies have shown that if a male and female dog are able to have a viable pregnancy, the offspring have a higher rate of survival. It is also important to note that the offspring are able to have puppies. This is important because it allows dogs to have kids of their own and increase the gene pool. }

Screenshot: https://ibb.co/Njg3BkC

So my money is on p-hacking. Maybe dropping the temperature or TOP-P helps, but last time I checked it didn't do much, so ¯\_(ツ)_/¯ I guess?

Re: Machine learning won't solve natural language understanding

#155

Post by qayxc » Tue, Aug 10, 2021, 4:01 PM UTC

moyix wrote: post #136

qayxc wrote: post #72

Earlier quoted context omitted.

There's a very important difference, though: humans can and often do qualify their interpretation. One major flaw of current PAC-based models is the inability to express certainty and being able to present the rationale behind their (lack of) confidence. One crucial aspect of human language processing is the ability to question or interrogate oneself to adjust one's interpretation. During a dialog this can be done by…

I find demos like yours hard to reason about because it is often the case that you can mess around with the prompt to get it to demonstrate understanding of things that it didn't seem to understand before. I wanted to see if it understood that humans and dogs can't have kids, and it does. Prompt is in [], generated text is in {}. ––––––––––––––––––––––––––––––––– [This is a question and answer series intended to demo…

Turns out dropping the temperature didn't help either, so yeah, p-hacking for the most part:

https://ibb.co/KNDyvsQ

Re: Machine learning won't solve natural language understanding

#156

Post by djoldman » Tue, Aug 10, 2021, 4:03 PM UTC

smallerfish wrote: post #129

djoldman wrote: post #76

"The man saw a boy with a telescope." There is NO exact interpretation of the above sentence. It is possible that the man peered through a telescope to see the boy, or that the man saw a boy who had a telescope. One may reasonably argue that one interpretation is more likely but either may be true. Ambiguity is an immense problem in semantic analysis, compounded by assumed speaker intent, etc. Ultimately there is no…

Right, but human conversation doesn't work like that. A real usage would be more like "The man saw a boy with his telescope. The boy was running down the beach, apparently screaming out at the sea." Using "a" in place of "his" in speech would be unnatural and would raise eyebrows. "His" is still ambiguous though, and any human speaker would know that and provide more context. The second sentence disambiguates the sen…

All good comments and I don't disagree completely. Most humans get around just find reading books and communicating.

Interestingly though, there are miscommunications between humans and not just because someone said "the man saw a boy with his telescope" without sufficient context.

The intractable problem is that human language has evolved to require context or presumption of intent and these are not stated or confirmed. Therefore probabilities MUST be assigned to them. There is no exactness because the context and intent are not stated/communicated/written, they are by definition assumed.

This assumed knowledge works in many cases but fails in others. The difficulty comes when the audience does not AGREE with the assumption.

TLDR: Humans communicate by assuming that their audiences share some of their beliefs but sometimes the unspoken beliefs of the audience don't match the speakers'.

Re: Machine learning won't solve natural language understanding

#157

Post by goatlover » Tue, Aug 10, 2021, 4:23 PM UTC

zR0x wrote: post #152

goatlover wrote: post #135

Earlier quoted context omitted.

Because other animals don't produce sounds which constitute abstract meaning that can be combined into sentences. Possibly a few birds and dolphins or whales might get close. But there's clearly something more going on in the brain than making sounds which contribute to word formation.

Sure there is more going on than we see on the surface. Before neuroscience and modern imaging we needed to sit and imagine and theorize. Now we don’t.

You don't think neuroscience involves theorizing? You think the neuroscientists can just image the neuronal activity and read it out in some complete scientific explanation?

Re: Machine learning won't solve natural language understanding

#158

Post by didibus » Tue, Aug 10, 2021, 4:53 PM UTC

darawk wrote: post #106

This article is....total nonsense. None of the challenges described are impossible to solve with statistical methods. And i'd be willing to bet nearly any amount of money that they will be solved that way, long before they are solved in any other way. The problems posed here are being addressed as we speak with knowledge graphs and graph neural networks, and some of the others can be addressed with more nuanced and c…

I don't think it's total nonsense, that seems a bit hyperbolic.

It makes some good points in that the amount of data it would take to approximate it all seems far too large, and if you look at a human baby, I'm not sure they are fed nearly as much data, yet they can develop language understanding.

I feel we have a missing piece here. Maybe it'll be solved by combining methods based on statically learning from data with others, but exclusively from learning on data might reach a limit which is we might not have enough data or computing power to do so.

Re: Machine learning won't solve natural language understanding

#159

Post by fishmaster » Tue, Aug 10, 2021, 5:00 PM UTC

canjobear wrote: post #39

The history of NLP is littered with people claiming on theoretical grounds that XYZ is unattainable using purely statistical methods, and that some notion of the logical structure of language needs to be brought in. And yet one by one, the XYZ have been attained by statistical methods. If you think there's something NLP can't do using machine learning, make a challenge dataset. That would be much more useful than yet…

There's the HANS dataset which is used to evaluate NLI models to check if they pick up surface heuristics: https://huggingface.co/datasets/hans https://arxiv.org/abs/1902.01007

Re: Machine learning won't solve natural language understanding

#160

Post by treesprite82 » Tue, Aug 10, 2021, 5:06 PM UTC

didibus wrote: post #158

darawk wrote: post #106

This article is....total nonsense. None of the challenges described are impossible to solve with statistical methods. And i'd be willing to bet nearly any amount of money that they will be solved that way, long before they are solved in any other way. The problems posed here are being addressed as we speak with knowledge graphs and graph neural networks, and some of the others can be addressed with more nuanced and c…

I don't think it's total nonsense, that seems a bit hyperbolic. It makes some good points in that the amount of data it would take to approximate it all seems far too large, and if you look at a human baby, I'm not sure they are fed nearly as much data, yet they can develop language understanding. I feel we have a missing piece here. Maybe it'll be solved by combining methods based on statically learning from data wi…

> It makes some good points in that the amount of data it would take to approximate it all seems far too large

The article's way of estimating this is absurd, seemingly relying on the idea that a model can't generalise and must see every possible variation of a sentence:

> If we add to the semantic differences all the minor syntactic differences to the above pattern (say changing ‘because’ to ‘although’ — which also changes the correct referent to “it”) then a rough calculation tells us a ML/Data-driven system would need to see something like 40,000,000 variations of the above