ChristianKl comments on Causal Diagrams and Causal Models - Less Wrong

61 Post author: Eliezer_Yudkowsky 12 October 2012 09:49PM

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

Comments (274)

You are viewing a single comment's thread.

Comment author: ChristianKl 14 October 2012 04:32:39PM 1 point [-]

An earthquake is 0.8 likely to set off your burglar alarm; a burglar is 0.9 likely to set off your burglar alarm. And - we can't compute this model fully without this info - the combination of a burglar and an earthquake is 0.95 likely to set off the alarm;

I don't think that true. The earthquake can cause the burgler to have less control over his own movement and therefore increase the chance that he triggers the alarm.

Comment author: Kindly 14 October 2012 06:10:33PM 1 point [-]

I don't think this matters too much to the main point, but if you like, you can imagine that with a 0.05 probability the alarm is incorrectly wired and will not go off no matter what happens.