When a base optimizer produces a learned model that is itself an optimizer, we want • outer alignment: the base optimizer does what the programmers want. but also • inner alignment: the learned optimizer does what the base optimizer wants. Analysis:https://www.alignmentforum.org/s/r9tYkB2a8Fp4DN8yB/p/pL56xPoniLvtMDQ4J …
0 replies
4 retweets
14 likes
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.