Peter_de_Blanc comments on Open Thread: March 2010 - Less Wrong

5 Post author: AdeleneDawner 01 March 2010 09:25AM

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

Comments (658)

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

Comment author: Peter_de_Blanc 02 March 2010 08:34:30PM 3 points [-]

One way to guess what might happen in a given situation is to compare it to similar situations in the past. Assume we already have some way of measuring similarity. Some past situations will be extremely similar to the current situation, and some will be less similar but still pretty close. How much weight should we attach to each?

If your data set is very small, then it is usually better for the weight to drop off slowly, while the opposite is true if your data set is large. Perhaps different individuals use different curves, and so some people will have an advantage at reasoning with scanty data, while others will have an advantage at reasoning with mountains of data. I thought that Alicorn was suggesting "luminosity" as a name for this personality trait. It looks like I was way off, though :-)