andreyk

45 ms·

511.
by andreyk 5y ago
"If you overwork yourself and feel like you’re burning out, stop what you’re doing and find stability—whatever you’re working on isn’t worth your health. Don’t feel the need to “hustle” because some people glorify it. Doing good work w
512.
Yannic Kilcher on Being an AI Researcher and Educator(thegradientpub.substack.com)
1 points by andreyk 5y ago|0 comments
513.
AI as a Hype Tool(skynettoday.com)
2 points by andreyk 5y ago|0 comments
514.
Using ML and Optimization to Solve DoorDash’s Dispatch Problem(doordash.engineering)
2 points by andreyk 5y ago|0 comments
515.
Staying Visual in a Clip World(liipetti.net)
1 points by andreyk 5y ago|0 comments
516.
AI as a Hype Tool(skynettoday.com)
5 points by andreyk 5y ago|1 comments
517.
by andreyk 5y ago
I think this is a bit too cynical/negative. It's not true that neural net approaches don't work, they just have different strengths and weaknesses (strength being generalization/flexibility/ease of use, weakness bei
518.
An Introduction to AI Story Generation(thegradient.pub)
10 points by andreyk 5y ago|0 comments
519.
Elon Musk Has No Idea What He’s Doing with Tesla Bot(spectrum.ieee.org)
58 points by andreyk 5y ago|86 comments
520.
Collaborative filtering doesn't work for us(about.chatroulette.com)
157 points by andreyk 5y ago|104 comments
521.
Alexander Veysov on Self-Teaching AI and Creating Open Speech-to-Text(thegradientpub.substack.com)
1 points by andreyk 5y ago|0 comments
522.
Systems for Machine Learning(thegradient.pub)
1 points by andreyk 5y ago|0 comments
523.
Systems for Machine Learning(thegradient.pub)
1 points by andreyk 5y ago|0 comments
524.
by andreyk 5y ago
Seriously, how do these kinds of things keep hitting first page HN? Quality aside, who on HN thinks this is of interest to the community in terms of being novel / worth discussing?
525.
by andreyk 5y ago
Thanks for letting us know!
526.
by andreyk 5y ago
True - this article presents a nice overview with more details on that https://www.nytimes.com/2020/01/12/technology/facial-recogni...
527.
by andreyk 5y ago
Hey HN! I am an editor on this piece, which I think highlights a trend not enough people are aware of. Would definitely welcome feedback / any knowledge wrt these kinds of news stories you think is relevant.
528.
by andreyk 5y ago
See 'Facial Recognition Should be Used with Care' - the police detectives should have been the 'human in the loop' in these cases, but appear to have done their job poorly in all these cases.
529.
Police Misuse of Facial Recognition – Three Wrongful Arrests and Counting(skynettoday.com)
51 points by andreyk 5y ago|16 comments
530.
by andreyk 5y ago
To be honest, I feel like Twitter, Reddit, Instagram, etc. are all a case of how you use it, it's not intrinsically that bad. And wrt short videos on TikTok, it's great that you enjoy the format, but I also think longer form video
531.
by andreyk 5y ago
"I see no reason that this technology couldn't smoothly scale into human-level intelligence, yet lots of people seem to think it'll require a step change or is impossible." I am a big fan of LMs and am not in the don&#x2
532.
by andreyk 5y ago
Oh man, I was looking forward to this a ton! Are you thinking to keep working on it with the open source GPT J or something similar by any chance?
533.
Machine learning won't solve natural language understanding(thegradient.pub)
195 points by andreyk 5y ago|188 comments
534.
by andreyk 5y ago
Not as well as I would have expected tbh - see the trint link above, it's pretty good but there are lots of errors, so correcting it is quite time consuming.
535.
by andreyk 5y ago
Good point! We should link to the papers discusses in the description at least, did not think of that.
536.
by andreyk 5y ago
Isbell's Neurips keynote is fantastic! Definitely recommend.
537.
by andreyk 5y ago
Just FYI, as I said in another comment in this case "many (senior) researchers agreed with Gebru's opposition to LeCun's original point - see tweets by Charles Isbel, yoavgo, Dirk Hovy embedded here https://thegrad
538.
by andreyk 5y ago
He means AGI is not a meaningful term, because 'general' is too ambiguous - ie, even human intelligence is not truly general. So it's a meaningless term according to him. I think he would not oppose a term such as 'super
539.
by andreyk 5y ago
Thanks! We'll work on that
540.
by andreyk 5y ago
Not to disagree, but a couple of FYIs: * LeCun did not really quit Twitter, he's still active on there and has been for a while - but I guess he did temporarily when all this happened. * many researchers agreed with Gebru's opposi

More ›