AlexMennen comments on Mathematical Measures of Optimization Power - Less Wrong

3 Post author: Alex_Altair 24 November 2012 10:55AM

You are viewing a comment permalink. View the original post to see all comments and the full post content.

Article Navigation

Comments (16)
Tags: utility expected_utility optimization_process optimization_power

Comments (16)

You are viewing a single comment's thread. Show more comments above.

Comment author: AlexMennen 25 November 2012 04:45:52AM 1 point [-]

So paralized people aren't less intelligent just because "raise arm" isn't an option for them (That is, no combination of action potentials in their head will cause their arm to move).

Caveat: if someone is paralyzed because of damage to their brain, rather than to their peripheral nerves or muscles, then this is not true, which creates and undesirable dependency of the measured optimization power on the location of the cause of the disability. Despite this drawback, I like this formalization.

Erm, you probably want to use something like (EU* - EU[av]) / EU[av], where EU* is just the actual expected utility, and EU[av] is the average of the expected utilities of the counterfactual probability distributions over world states associated with each of the agents options.

No, that clearly makes no sense if EU[av] <= 0. If you want to divide by something to normalize the measured optimization power (so that multiplying the utility function by a constant doesn't change the optimization power), the standard deviation of the expected utilities of the counterfactual probability distributions over world states associated with each of the agent's options would be a better choice.

Comment author: aaronde 25 November 2012 06:05:31AM 1 point [-]

Caveat: if someone is paralyzed because of damage to their brain, rather than to their peripheral nerves or muscles, then this is not true,

That's why I specified that the you don't get penalized for disabilities that have nothing to do with the signals leaving your brain.

which creates and undesirable dependency of the measured optimization power on the location of the cause of the disability.

I disagree. I think that's kind of the point of defining "optimization power" as distinct from "power". A man in a prison cell isn't less intelligent just because he has less freedom.

No, that clearly makes no sense if EU[av] <= 0. If you want to divide by something to normalize the measured optimization power (so that multiplying the utility function by a constant doesn't change the optimization power), the standard deviation of the expected utilities of the counterfactual probability distributions over world states associated with each of the agent's options would be a better choice.

Great idea! I was really sloppy about that, realized at the last minute that taking a ratio was clearly wrong, and just wanted to make sure that you couldn't get different answers by scaling the utility function. I guess |EU[av]| does that, but now we can get different answers by shifting the utility function, which shouldn't matter either. Standard deviation is infinitely better.

Powered by Reddit Powered by Reddit