5 - Infra-Bayesianism with Vanessa Kosoy
AXRP - the AI X-risk Research Podcast
00:00
Is the State of Infinite Utility a Good Predictor?
If a prediction turns out to be false, we can imagine that what happens is a transition to some state of infinite utility. This is consistent with evations, because the agent will never see a situation in which a prediction is falsified. So it is consistent, from the agent's perspective, to assume this. But once we assume this, then the optimo policy become behave as if those predictions will be true. And our policy is going to be the optimal policy given that omega predicts correctly, which is the unity policy. Now, there is another thing where you can develop it. If you also allow o conficsets which are empty,. so cativ believes it can have described a
Transcript
Play full episode
