Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
purplecpa2
on Feb 10, 2017
|
parent
|
context
|
favorite
| on:
DeepMind AI learns to act aggressive when it doesn...
The AI acted aggressively by blasting its opponents....because the game was designed with that being the optimal way to score points. If the optimal way to win was to just pause the game, it would have done that. 0/5 stars, would not read again.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search: