Multi-dimensional rewards for AGI interpretability and control

📖 Description

*Update August 2021:* Re-reading this post, I continue to think this is a good and important idea, and I was very happy to learn after I wrote it that what I had in mind here is really a plausible, viable thing to do, even given the cost and performance requirements that people will demand of our future AGIs. I base that belief on the fact that (I now think) the brain does more-or-less exactly what I talk about here (see my post [A model of decision-making in the brain](https://www.lesswrong.com/posts/e5duEqhAhurT8tCyr/a-model-of-decision-making-in-the-brain-the-short-version)), and also on the fact that the machine learning literature also has things like this (see the comments section at the bottom).

📊 Game Impacts

💭 Reactions

🔗 Sources

• [1] https://www.alignmentforum.org/posts/Lj9QXcqkcuR4iHJ7Q/multi-dimensional-rewards-for-agi-interpretability-and