RobinZ comments on Causation as Bias (sort of) - Less Wrong

-12 Post author: spuckblase 10 July 2009 08:38AM

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

Comments (88)

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

Comment author: RobinZ 10 July 2009 05:53:21PM 2 points [-]

The model is less complex, possibly, but this is not the sole criterion on which it should be judged. Occam's Razor discourages complexity, but it encourages precision, and a Humeiform universe avoids the first only at the cost of utterly failing the second.

To use a metaphor: you have to pay for everything you put in your model, but your return on that investment is related to how much accuracy the model gives. More expensive (complex) models can still win if they give a bigger enough payoff (sufficiently complete prediction of the observations).