[YeGoblynQueenne]

>> My comments on understanding can be summarized as such: either

Please don't do this. Too many assumptions about what and how I think leave a bad taste.

Yes, a model that can identify synonyms accurately lacks human faculties, including understanding. That's what modern machine learning boils down to. There are many tasks we thought would require human intelligence or reasoning, that can, after all, be reduced to dumb classification. In other words, there is no need to claim "understanding" to explain the output of a classification model, just because a human can perform the same task _and_ can understand it.

As to the representation- that is the only thing that matters. If you want to claim a model represents a concept, you have to be able to show where in the model's structure that concept is represented. If there is a representation- where is it?

[joshuamorton]

This response runs much to close to my position number one for me to have any more interest in continuing. This line of discussion. I'll just leave you again with one question

> If there is a representation- where is it?

Where is your understanding of language?

[YeGoblynQueenne]

(HN thinks we've overdit it. This goes to your comment below).

But, I'm a human being. Why do I need to show you my representation to convince you I possess human understanding of language?

Conversely, to claim that a statistical model possesses understanding is a very strong claim that requires equally strong evidence. And since we can inspect a statistical model's representation- that is where the evidence should be sought.

[YeGoblynQueenne]

Why does that matter? Is there any doubt as to whether I can understand language? The question is whether word2vec etc can.

[joshuamorton]

>Is there any doubt as to whether I can understand language?

According to your own metric, yes! You, not I, are the one who claimed

> If you want to claim a model represents a concept, you have to be able to show where in the model's structure that concept is represented.

So by your own metric, I must question whether or not you understand language, since you have yet to point out to me where you represent that understanding.

(To be clear, I think you absolutely do understand language, but I think the whole idea that the structure matters is absurd. It clearly doesn't to you or I, so why hold a statistical model to a higher standard of "understanding" than you hold yourself to?)

[YeGoblynQueenne]

(also posted above- wasn't getting a reply link).

But, I'm a human being. Why do I need to show you my representation to convince you I possess human understanding of language?

Conversely, to claim that a statistical model possesses understanding is a very strong claim that requires equally strong evidence. And since we can inspect a statistical model's representation- that is where the evidence should be sought.

reply

[joshuamorton]

>Why do I need to show you my representation to convince you I possess human understanding of language?

Why not? What makes you special? You're giving humans a double standard here. It's not like we have a particularly clear understanding of the human model of cognition.

And since we're venturing further into the philosophical, I feel safe making the assertion that no, I, a priori, have no reason to believe that you have more understanding of language than a model. I'm only interacting with you online. For all I know, you are a computer program. Why should I trust you and hold you to a different standard than I would any other model?

[YeGoblynQueenne]

>> You're giving humans a double standard here.

There is no double standard and the matter is not philosophical. We know humans to be able to understand language- we do not need to prove it in any way, including by examining our representation of meaning in language.

We don't know that statistical language models have an understanding of language, it's a very strong assumption to make that they do, and it must be justified with equally strong evidence. Their internal structure is available for inspection, therefore if they are capable of understanding language this understanding should have a concrete reprsentation that we can identify. If we can't, then they don't have anything like "understanding".

>> Why should I trust you and hold you to a different standard than I would any other model?

Assuming that I'm not a model is a reasonable assumption with a very high probability of being true and the simplest explanation for my participation in this thread.