I aspire to perfectionism, but you can’t go from there to my thinking that any given system is already perfect. Especially not evolution!
As for black swans, you need more cognitive complexity, not less, to handle them; Gaussian randomness is easy by comparison. Noise doesn’t help with black swans; a random key does not fit a random lock. Evolution in particular does very poorly with black swans. All this will be one or more separate posts at some point.
Rooney, the difference is between a qualitative view and a quantitative view. If you assign a 90% probability to one belief and a 10% probability for another, you can search for a plan with a decent expected utility—a decent weighted combination of probable results given both beliefs. That really isn’t the same as believing some things, and not believing others. It’s a distinction with a difference.
I aspire to perfectionism, but you can’t go from there to my thinking that any given system is already perfect. Especially not evolution!
As for black swans, you need more cognitive complexity, not less, to handle them; Gaussian randomness is easy by comparison. Noise doesn’t help with black swans; a random key does not fit a random lock. Evolution in particular does very poorly with black swans. All this will be one or more separate posts at some point.
Rooney, the difference is between a qualitative view and a quantitative view. If you assign a 90% probability to one belief and a 10% probability for another, you can search for a plan with a decent expected utility—a decent weighted combination of probable results given both beliefs. That really isn’t the same as believing some things, and not believing others. It’s a distinction with a difference.