LESSWRONG
LW

Outer AlignmentAI

1

Cooperative Game Theory

by [anonymous]
7th Jun 2023
1 min read
0

1

This post was rejected

Outer AlignmentAI

1

New Comment
Moderation Log
Curated and popular this week
0Comments

Just spit balling ideas.  Similar to PINNs

Informing loss function with cooperative game theory

Could at least hope for lower level "peace oracles" 

Aiming for cooperation with any actors encountered  

Center it in a system like hugging GPT

RLHF at end