Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Shirt Without Stripes

github.com

Post reply on HN
Print view View original

471–480 of 639 posts

Re: Shirt Without Stripes

#471

Post by animalCrax0rz »

larl wrote: post #430
animalCrax0rz wrote: post #411

Earlier quoted context omitted.

Google is in the business of producing quick results to sell ads while keeping the cost low. If anyone does better, they will likely do it using a costlier algorithm, and if there is more margin in reselling tangible products than ads then Amazon is incentivized to use costlier algorithms that are more accurate. But I think the margin in both cases does not justify the use of algorithms that may be more accurate in s…

animalCrax0rx is in the business of producing quick pithy posts to earn more karma while keeping understanding low and making claims that are not evidence based... See how that works? That's not really what's going on. Sure, G. is incentivized to include pages quickly, but they are also incentivized to produce them accurately, and as the above poster indicates, this is quite a hard problem to solve generally. A is al…

Google is aware of this problem in their search approach. It's a business problem, not a technical one. You're saying the same thing in suggesting they base their decisions on some measurement set. If solving the problem adds complexity, which it certainly will, and there is not enough improvement in accuracy for the majority of cases in their measurement set, why bother? You sound like the kind of person that attacks people for their opinion. So weird, dude.

Re: Shirt Without Stripes

#472

Post by animalCrax0rz »

atq2119 wrote: post #451
animalCrax0rz wrote: post #314

Earlier quoted context omitted.

Which part of "The paper I linked to below describes a technique they call "deep parsing." Check it out for more context." could you not parse? I mean if us humans have a difficulty parsing each other's statements, then why should machines do any better?

None. I was merely pointing out that your comment wasn't a response to albertzeyer's point about statistical methods. That is to say, maybe you didn't parse their comment properly ;)

Nope, I was simply countering that it's not as simple as they suggested. I added the link to the paper to show one approach to given them some context as to what might it look like.

Re: Shirt Without Stripes

#473

Post by dang »

aerovistae wrote: post #123

This comment would go from "unreadable" to "interesting" if you had phrased it as: "Vaguely similar to a joke from _the movie_ Ninotchka that _the Slovenian philosopher_ Zizek often uses...." Give people context. Don't assume people know what you know.

We detached this subthread from https://news.ycombinator.com/item?id=22925490 and marked it off topic.

Re: Shirt Without Stripes

#474

Post by ddlutz »

Noticed something interesting, if you search for 'shirt without sleeves' in google images, you DO get sleeveless shirts. So why doesn't this work with stripes?

Re: Shirt Without Stripes

#476

Post by Polylactic_acid »

wonderwonder wrote: post #238
Alex3917 wrote: post #212

Earlier quoted context omitted.

At least this is relatively innocuous. Until recently if you did a Google Image Search for "person" or "people", it only showed white men.

Most of the person results appear to be 'Time Person of the Year' related. Another result is a guy with the last name Person. The results don't seem to be related to the definition of the word 'person'.

For me it shows all newsworthy people and articles. It shows the titles of the pages and they are all stuff like "11 signs you are a good person" So it seems clear that there is no kind of AI bias here but simply that high ranking articles with the word person more often than not choose white men as their stock image.

Most of the very top results seem to be of trump and greta thunberg.

Re: Shirt Without Stripes

#477

Post by BickNowstrom »

albertzeyer wrote: post #243

Why should it not be possible to solve this with statistical methods? The model just needs to be able to understand the important meaning of "no" in here, in the context of the whole sentence. I would guess that most modern NNs from the NLP area (Transformer or LSTM) would be able to correctly differentiate the meaning. The problem is, I think there is no fancy NN (yet) behind Google search, and the other web searche…

> Why should it not be possible to solve this with statistical methods?

Doing just that for 10 years, beating hand-coded systems: https://www-nlp.stanford.edu/pubs/SocherLinNgManning_ICML201... [pdf]

> I would guess that most modern NNs from the NLP area (Transformer or LSTM) would be able to correctly differentiate the meaning.

Yes. See demos like: https://demo.allennlp.org/constituency-parsing/MTczNjYyNA== and https://demo.allennlp.org/dependency-parsing/MTczNjYyNg==

> I think there is no fancy NN (yet) behind Google search,

During the deep learning boom, Google made a huge push towards NN-based NLP. SEO's and their PR calls their efforts collectively RankBrain: https://en.wikipedia.org/wiki/RankBrain

I think we are on the cusp of combining symbolical/logical operations over the vectors produced by Neural Networks (or at least, major effort there). Could be by neatly tying up all these different NN-based NLP modules (parsing, semantic distance, knowledge bases, ...) with another set of decision layers stacked on top.

Re: Shirt Without Stripes

#478

Post by ltbarcly3 »

Here's another fun fact about how commerce search engines work (I spent a couple of years on this):

Negations sidestep almost all of the algorithms that try to provide an improved result set, and fall through to pure text relevancy. So try searching on amazon for shirt, then search for: shirt -xkxkxkxk. Since xkxkxkxk doesn't match any documents, the negation should have no effect, but it does, the effect it has is to sidestep all the fancy relevancy work and hardcoded query rewrite rules, domcat rules, demand and sales/impression statistics etcetc, and give you basically awful search results. You don't even get shirts.

Re: Shirt Without Stripes

#479

Post by cfv »

hbosch wrote: post #466
cfv wrote: post #464

While I'm sure this is A Hard Problem to solve by NLP I for whatever reason was under the impression that this is trivial to special-case. As in, "X without Y" sounds like a common enough use case to have it's own little parser branch in places as big as Google or Amazon

I mean, if I google the phrase "shirts -stripes" and click the Images tab I see mainly shirts without stripes. So it's essentially the same input, and essentially the same expected output, but there must be quite a knot between understanding the word "without" and literally just using the - operator.

Right, but the - operator requires prior knowledge that the search engine understands boolean operators and that a -b is an alias of "listings with a on their text that don't have b in their text too" whereas "a without b" is inmediately recognizable by whoever is writing the search as "I want something of kind A without property B"

Re: Shirt Without Stripes

#480

Post by nailer »

larl wrote: post #430
animalCrax0rz wrote: post #411

Earlier quoted context omitted.

Google is in the business of producing quick results to sell ads while keeping the cost low. If anyone does better, they will likely do it using a costlier algorithm, and if there is more margin in reselling tangible products than ads then Amazon is incentivized to use costlier algorithms that are more accurate. But I think the margin in both cases does not justify the use of algorithms that may be more accurate in s…

animalCrax0rx is in the business of producing quick pithy posts to earn more karma while keeping understanding low and making claims that are not evidence based... See how that works? That's not really what's going on. Sure, G. is incentivized to include pages quickly, but they are also incentivized to produce them accurately, and as the above poster indicates, this is quite a hard problem to solve generally. A is al…

Your ad hominem on the other poster is unnecessary, violates the HN guidelines and not an apt comparison.

Pointing out the obvious: Google is an advertising company. If the cost of producing an accurate result outweighs the advertising income on a given term, there is no incentive for Google to produce better results.

Post reply on HN