Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
timbre
on Nov 21, 2013
|
parent
|
context
|
favorite
| on:
High Frequency Dating
It's really the wrong approach. Supervised learning is the way to go. For example a paper by Kumar et al.[1] shows how to build an "attractive woman" classifier that is 83% accurate.
[1]
http://homes.cs.washington.edu/~neeraj/publications/base/pap...
anaphor
on Nov 21, 2013
|
next
[–]
Presumably if they used 310 million faces instead of only 3.1 million it would be even more accurate, which is pretty impressive.
VladRussian2
on Nov 21, 2013
|
prev
[–]
i'm confused - an .edu paper has been already mentioned while "Weird Science" hasn't been yet.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
[1] http://homes.cs.washington.edu/~neeraj/publications/base/pap...