HackerTrans
トップ新着トレンドコメント過去質問紹介求人

anjc

no profile record

コメント

anjc
·2 か月前·議論
Your litmus test isn't correct and your assumption of personalisation isn't correct either. All of the criteria that you see as fine are controlled under the relevant legislation and are considered personalisation, requiring transparency etc.

Furthermore, bills have been brought to EU parliaments that have erroneously attempted to ban all forms of ranking, which would include even the most basic information retrieval algorithms. So it isn't obvious at all what is meant by 'algorithm'.
anjc
·6 年前·議論
RecSys 2016

Covington, Paul, Jay Adams, and Emre Sargin. "Deep neural networks for youtube recommendations." Proceedings of the 10th ACM conference on recommender systems. 2016.

Nvidia furthermore implemented/released it for TensorFlow, and for their new Recsys engine:

https://ngc.nvidia.com/catalog/resources/nvidia:wideanddeep_...
anjc
·6 年前·議論
I'm sorry I don't know how to turn it on. When I go to youtube.com, at the top it has about 15 categories that it has gleaned from my use (e.g. Guitar, Flight Simulators), some of which I'm surprised it learned, as well as All, Mix, Trending. I use Edge and I'm logged in to YouTube.
anjc
·6 年前·議論
> Then, if a user watches a single video on a topic outside of their usual interest scope, YouTube replaces any semi-relevant recommendations with with that new category, regardless of if the user is actually interested in the topic.

That's the way it used to work, now it adds that new category to the list of categories it thinks you like (top of the home page), and you can filter or include that category.
anjc
·6 年前·議論
This criticism shouldn't apply anymore. They predict categories you like and you can select a category. The 'All' page should be a mix of all categories, not just cat videos. Amazon on the other hand...
anjc
·6 年前·議論
> Because YouTube doesn't care what you want--YouTube cares what it can monetize.

> You are the product, not the customer.

> Once you internalize that, stuff that Google does makes sense.

They published their algorithm. They rank items according to predicted watch-time. The other poster is probably correct, in that many people use these long classical music videos as background music over and over (i.e. high watch time) and so they're highly ranked.
anjc
·6 年前·議論
If they used to recommend a good mix of content, what do you feel is the issue with recommendations now? Not enough variety? Not enough relevance?
anjc
·6 年前·議論
Since when, do you think? I personally think that they're better than ever. On the other hand, I think they moved to more deep learning approaches over the last 3-5 years (based on their research output at least) so perhaps recommendations are perceivably different since then
anjc
·6 年前·議論
If they were only recommending comedy from the same artist you may then have the same issue as another comment, which is that recommendations aren't novel enough.

I'd take your point if they only recommended bass videos, but presumably it's just one recommendation in a list of many, with the point being to explore your preferences while also satisfying your demands
anjc
·6 年前·議論
I'm always mind-blown reading comment threads like this regarding recommendation, with people annoyed that explicit feedback helps recommenders, or exasperated that they're recommended one item that they haven't approved while also wanting novel recommendations.

1000s of hours of videos are uploaded every minute. 5+ billion videos are watched daily. Are you not rather quite pleased that YouTube can deliver highly relevant top-20 recommendations within milliseconds from such a large and changing corpus, while keeping track of your past preferences, your interactions, your social circle's preferences, popularity and trends, content features, etc?

It's pretty amazing to me. I think people expect recommender systems to read minds.
anjc
·6 年前·議論
> "use YouTube infrequently"

> "binging random content"

gets recommended random popular content

> "why are recommendations so bad"

They can't read your mind, you know
anjc
·9 年前·議論
He meant Seikos&Pulsars
anjc
·9 年前·議論
I get what you're saying. But I'd argue that it isn't the watches themselves which scream wealth, but rather the brand. Well perhaps being able to say that your watch is made from solid gold would signify a level of wealth too :)
anjc
·9 年前·議論
Yeah I agree with that. Although that's arguably similar for many high-end brands. I certainly know of Chanel, say, and even know their general history somehow, but I wouldn't recognise their products in a line-up with clothes from the cheapest high street stores.
anjc
·9 年前·議論
Automatics do sweep though right? Rolex do anyway

Could be a good project if you could make headway
anjc
·9 年前·議論
I think you're misinterpreting the mass market. Yes if you're into watches you know Patek. How many people are into watches? Even the people who can afford to buy Pateks aren't necessarily into watches. I know nobody who would know them.

On the other hand, Rolex is literally one of the most recognised and reputable brands in the world. There are companies which track this stuff, and every study ranks them within top 100 in the world for brand equity, recognition, reputation, etc among general consumers. Just with a cursory look now, I see one study from 2014 which ranks them second in the world, only behind Disney:

http://www.forbes.com/sites/arieladams/2014/04/14/why-rolex-...

Don't forget, they sponsor a lot of stuff. You can't watch sports without seeing their logo everywhere. As I said, people wouldn't recognise a Rolex watch...but the brand, yeah.

I'd be extremely surprised if 90% of people in a bank in Switzerland don't know of the brand Rolex.
anjc
·9 年前·議論
I think Seiko 5's are special given the price and the standards adhered to.
anjc
·9 年前·議論
I don't know anything about movements other than what I can see through transparent backs and Wikipedia articles, but...they look sensitive to experimentation
anjc
·9 年前·議論
Even a good fake with a Swiss movement could cost $500-1000. It might still be worth robbing.
anjc
·9 年前·議論
>>"Can't tell if this post is for real, but either way I respectfully disagree. Ever worn a Rolex? 99% of people won't recognize it. From those ppl that do recognize it and know the price, some will think its expensive and some will think its quite inexpensive. Rolex sells almost a million watches a year now. It's a brand bought by rich, knowledgeable watch nerds as well as plebeian, flashy status-seekers and everyone in between. Rolex are given as gifts often enough whether from family or from your old fat boss for you being the best salesman in the district."

You're disagreeing with yourself and agreeing with him. His point (I think) is that they adopted a marketing and pricing strategy which promoted brand awareness and maintained brand equity to both the "lower classes", and to the "upper classes" and also to watch enthusiasts, simultaneously.

That is to say, poorer people might recognise Rolexes as being something special, and might strive to buy one, but wealthier people also see it as something special and strive to buy one. And the fact that poorer people can get them doesn't dilute their positioning for wealthier demographics. Or as you put it..."It's a brand bought by rich, knowledgeable watch nerds as well as plebeian, flashy status-seekers and everyone in between."

And I completely disagree with your first statement. I would imagine that almost every adult recognises the Rolex brand. Sure, people might not recognise a Submariner from a distance as being a Rolex, or a Rolex design, but everybody knows the name. Whereas, on the other hand, almost nobody would even know the name Philippe Patek. But then, that's their marketing strategy.