Shirt Without Stripes

Post by acdha » Tue, Apr 21, 2020, 1:23 PM UTC

fyfy18 wrote: post #521

acdha wrote: post #514

Earlier quoted context omitted.

Google has for years put out puff pieces talking about high accuracy on image tagging. It’s only within the last few months that searching my Photos library for “cat” returned something other than pictures of my dog. There’s a nuanced argument that practitioners know how ML is so dependent on training data and accuracy tails off sharply, but that nuance tends to removed from anything selling to potential customers —…

I'd assume pets are hard as there are so many varieties (potentially even harder than humans). For the last few years Google Photos has correctly returned photos for a search of "Lamborghini" in my albums. I'd expect "shirt stripes" to fall into that category.

Post by otikik » Tue, Apr 21, 2020, 1:27 PM UTC

Post by nailer » Tue, Apr 21, 2020, 1:59 PM UTC

jefftk wrote: post #492

nailer wrote: post #480

Earlier quoted context omitted.

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.

This would predict that a query that has no advertising income will return no results, which is clearly not the case. Having a search engine that people go to whenever they want to search for things is incredibly valuable, because they will come to you when they want to buy things and you can sell ads. But unless you consistently give the best results for all queries, people will go whenever does. It's worth investin…

Post by DarkWiiPlayer » Tue, Apr 21, 2020, 2:28 PM UTC

jedberg wrote: post #390

bartread wrote: post #221

Earlier quoted context omitted.

I couldn't quite believe your comment when I read it so I did a Google image search for "person" and the results weren't a lot better than you'd suggested. Mostly white men, a few white women, a very few black women, a handful of Asians, and multiple instances of Terry Crews. The net result of that Google search, combined with the "Shirt Without Stripes" repo, leaves me even more unimpressed with the capabilities of…

If you really want to be disappointed, search for [doctor] and [nurse]. Unless things have really changed, [doctor] will be mostly white men and [nurse] will be mostly white and Filipino women. But don't blame the AI. The AI has no morality. It simply reflects and amplifies the morality of the data it was given. And in this case the data is the entirety of human knowledge that Google knows about. So really you can't…

Post by drongoking » Tue, Apr 21, 2020, 3:09 PM UTC

ghayes wrote: post #547

zaat wrote: post #494

Earlier quoted context omitted.

Shirt without paisley - fails shirt without buttons - preety much fail. shirt without red button - as already expected, shirts with red buttons

To be fair, the first and last query would also confuse me if someone asked me for that item. "Shirt without paisley" feels a bit like "cereal without elephants." You don't usually explicitly exclude an element that is relatively uncommon.