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

Jed_Harris comments on Tsuyoku Naritai! (I Want To Become Stronger) - Less Wrong

111 Post author: Eliezer_Yudkowsky 27 March 2007 05:49PM

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

Comments (73)

Sort By: Old

You are viewing a single comment's thread.

Comment author: Jed_Harris 14 April 2007 07:21:24PM 0 points [-]

Great discussion! Regarding majoritarianism and markets, they are both specific judgment aggregation mechanisms with specific domains of application. We need a general theory of judgment aggregation but I don't know if there are any under development.

In a purely speculative market (i.e. no consumption, just looking to maximize return) prices reflect majoritarian averages, weighted by endowment. Of course endowments change over time based on how good or lucky an investor is, so there is some intrinsic reputation effect. Also, investors can go bankrupt, which is an extreme reputation effect. If investors reproduce you can get a pretty "smart" system, but I'm sure it has systematic limitations -- the need to understand those limitations is a good example of why we need a general theory of judgment aggregation.

I'd like to see an iterated jelly bean guessing game, with the individual guesses weighted by previous accuracy of each individual. I bet the results would quickly get better than just a flat average. Note that (unlike economies) there's no fixed quantity of "weight" here. Conserved exchanges are not a necessary part of this kind of aggregation.

On the other hand if you let individuals see each other's guesses, I bet accuracy would get worse. (This is more similar to markets.) The problem is that there's be herding effects, which are individually rational (for guessers trying to maximize their score) but which on average reduce the overall quality of judgment. This is an intrinsic problem with markets. Maybe we should see this as an example of Eliezer's point in another post about marginal zero-sum competition.