News for hackers. Stuff that matters.

A comment on Apple contractors listened to 1k Siri recordings per shift: former employee. Parent comment

Re:Apple contractors listened to 1k Siri recordings per shif... by SpaceRaccoon on Saturday August 24, 2019 @06:32PM (#20790131)
> Well obviously! A machine can't train itself!

In some cases it can. That's what unsupervised learning is.

[ Parent | on HN ]
Re:Apple contractors listened to 1k Siri recordings per shif... by kentonv on Saturday August 24, 2019 @06:45PM (#20790172)
See that's sort of what I thought but now I'm being told it was silly to expect that. ¯\_(ツ)_/¯
[ Parent | on HN ]
Re:Apple contractors listened to 1k Siri recordings per shif... by wolfi1 on Saturday August 24, 2019 @06:46PM (#20790177)
one expert in ML told me once: unsupervised ML is a myth. I tend to believe him.
[ Parent | on HN ]
Re:Apple contractors listened to 1k Siri recordings per shif... by tormeh on Saturday August 24, 2019 @06:55PM (#20790209)
It's better to say that unsupervised ML is merely indirectly supervised ML.
[ Parent | on HN ]
Re:Apple contractors listened to 1k Siri recordings per shif... by PhilippGille on Sunday August 25, 2019 @06:01AM (#20792182)
Wasn't there a second version of AlphaGo that was trained only by playing against itself without external input and turned out to be even better than the first version that was trained with external input?

Isn't that unsupervised learning?

[ Parent | on HN ]
Re:Apple contractors listened to 1k Siri recordings per shif... by skj on Sunday August 25, 2019 @08:23AM (#20792562)
You have ground truth on the result: one agent won the game.

With voice recognition, that's a lot more vague and human supervision is helpful.

[ Parent | on HN ]
Re:Apple contractors listened to 1k Siri recordings per shif... by dekhn on Sunday August 25, 2019 @11:37AM (#20793524)
the rules of the game, and the output from winning, is supervision.
[ Parent | on HN ]