Steven Byrnes comments on [Intro to brain-like-AGI safety] 5. The “long-term predictor”, and TD learning

[Steven Byrnes]

Here’s a twitter thread wherein Nathaniel Daw gently pushes back on my dopamine neuron discussion of Section 5.5.6.

EDIT MUCH LATER TO ADD: As noted in the changelog, I wound up deciding that he was right, and heavily editing that part of the post.