Hacker News
new
|
past
|
comments
|
ask
|
show
|
jobs
|
submit
login
smokel
on April 7, 2025
|
parent
|
context
|
favorite
| on:
DeepMind program finds diamonds in Minecraft witho...
You must be confused. This research is about reinforcement learning, not about large language models.
ninetyninenine
on April 7, 2025
[–]
It's parroting human reinforcement.
danijar
on April 7, 2025
|
parent
[–]
It actually has no human data as input and learns by itself in the environment, that's the point of the accomplishment! :)
ninetyninenine
on April 8, 2025
|
root
|
parent
[–]
That's what humans do right? So it's parroting us.
Guidelines
|
FAQ
|
Lists
|
API
|
Security
|
Legal
|
Apply to YC
|
Contact
Search:
