[SEQ RERUN] Measuring Optimization Power
Today’s post, Measuring Optimization Power was originally published on 27 October 2008. A summary (taken from the LW wiki):
In order to measure the power of an optimization process, we can calculate how improbable it is that its goals would be fulfilled if that process were not present. The more unlikely they are, the more powerful the process is.
Discuss the post here (rather than in the comments to the original post).
This post is part of the Rerunning the Sequences series, where we’ll be going through Eliezer Yudkowsky’s old posts in order so that people who are interested can (re-)read and discuss them. The previous post was Aiming at the Target, and you can use the sequence_reruns tag or rss feed to follow the rest of the series.
Sequence reruns are a community-driven effort. You can participate by re-reading the sequence post, discussing it here, posting the next day’s sequence reruns post, or summarizing forthcoming articles on the wiki. Go here for more details, or to have meta discussions about the Rerunning the Sequences series.
If you’re going to numerically measure the optimisation power of an agent, then rather than its performance on a one-off problem, shouldn’t it be a rate of bits per second rather than a fixed number of bits?
Except that that makes it proportional to clock speed, and you can’t make a desk calculator intelligent by speeding it up a billionfold.
Problems with this approach have been discussed here.