Brendan
9
2
Brendan has not written any posts yet.
Brendan has not written any posts yet.
"For a true Bayesian, information would never have negative expected utility". I'm probably being a technicality bitch, attacking an unintended interpretation, but I can see bland examples of this being false if taken literally: A robot scans people to see how much knowledge they have and harms them more if they have more knowledge, leading to a potential for negative utility given more knowledge.
When I read Eleizer's posts on free will, and then spent time thinking it through myself, I came to the conclusion that the question was (non-obviously) incoherent, and that this is what Eleizer also thinks. More specifically, that when you Taboo free will you find yourself trying to say that your brain is not controlled by physics, rather "you" do, where "you" has to be not physics, which is really just "magic". I started thinking further about questions and realised that there are whole classes of questions for which neither "Yes" or "No" is the correct response. The most basic example is a contradiction. P = A is true & A... (read more)