Less Wrong is a community blog devoted to refining the art of human rationality. Please visit our About page for more information.

cousin_it comments on The Optimizer's Curse and How to Beat It - Less Wrong

44 Post author: lukeprog 16 September 2011 02:46AM

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

Comments (81)

You are viewing a single comment's thread.

Comment author: cousin_it 17 September 2011 02:10:09PM *  7 points [-]

Am I missing something, or does the post just say that we shouldn't use frequentist "unbiased estimators" as if they were Bayesian posterior expected values?

Comment author: jsalvatier 17 September 2011 03:48:59PM 4 points [-]

Not quite. If you were to do individual bayesian estimates you would have the same problem because there is shared prior information that would remain unmodeled.

Comment author: cousin_it 17 September 2011 08:05:54PM *  5 points [-]

Are you pointing out that each individual Bayesian estimate must be conditioned on all the information available, or is it more subtle than that?

Comment author: jsalvatier 17 September 2011 11:31:36PM 3 points [-]

Nope, that's it.