Ah, so chopping off some employees makes for a great team... You remove useless
data and get the same thing anyhow once decompressed. The model for employee
pattern learning learns patterns and can add back employees correctly in
diverse businesses. Wait insn't that a GPT-2 Task to say who is who and who is
added? The shrunken business of REAL humans is just less data in their noggins,
it IS about knowledge still! You remove bad employees based on useless wisdom
they think they know. And the team works better and does the same thing, just
more efficiently. They are/ know general knowledge and patterns that solve all
sorts of unseen issues. So is your suggestion not to compress text, but to
remove useless brains (holding useless data)? I think that already happens but
I wouldn't try it myself. It requires killing.
------------------------------------------
Artificial General Intelligence List: AGI
Permalink:
https://agi.topicbox.com/groups/agi/T26a5f8008aa0b4f8-Ma204a62e2737e67cb0b2402d
Delivery options: https://agi.topicbox.com/groups/agi/subscription
