Hacker News .hn
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
nl
on Sept 28, 2016
|
parent
|
context
|
favorite
| on:
Google’s Neural Machine Translation System
I think you are overestimating human performance here.
The weakness in the evaluation system is mostly because humans are imperfect translators.
nomailing
on Sept 28, 2016
[–]
That is why there are metrics like inter rater reliability and significance tests and confidence intervals, which seem to be all missing in paper...
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
The weakness in the evaluation system is mostly because humans are imperfect translators.