"When it's not overly costly, you should deviate from your usual theories of optimal behavior for the sake of expected information gain. Even in theory, empiricism is necessary... For a Bayesian optimizing agent in an uncertain world, information has positive expected utility, and experiments have positive expected information. Ergo, do them sometimes!" (This good idea is from Less Wrong blog. The pic from SFGate and for some reason it's titled The Little Black Dress Experiment)
No comments:
Post a Comment