wu :: forums
« wu :: forums - Bayesian Truth Serum »

Welcome, Guest. Please Login or Register.
Nov 28th, 2024, 2:48pm

RIDDLES SITE WRITE MATH! Home Home Help Help Search Search Members Members Login Login Register Register
   wu :: forums
   riddles
   general problem-solving / chatting / whatever
(Moderators: Eigenray, towr, SMQ, ThudnBlunder, Grimbal, william wu, Icarus)
   Bayesian Truth Serum
« Previous topic | Next topic »
Pages: 1  Reply Reply Notify of replies Notify of replies Send Topic Send Topic Print Print
   Author  Topic: Bayesian Truth Serum  (Read 559 times)
amichail
Senior Riddler
****





   


Posts: 450
Bayesian Truth Serum  
« on: Oct 20th, 2006, 9:00pm »
Quote Quote Modify Modify

What do you think of this?
 
http://targetyournews.com/?cmd=surf&urlid=656255
IP Logged

DropZap - a new kind of block elimination game
Whiskey Tango Foxtrot
Uberpuzzler
*****



Sorry Goose, it's time to buzz a tower.

   
Email

Gender: male
Posts: 1672
Re: Bayesian Truth Serum  
« Reply #1 on: Oct 20th, 2006, 9:15pm »
Quote Quote Modify Modify

If I am reading this correctly, he is using truth serum to establish a basis for evaluating a subjective opinion based on pre-established statistical expectations.
 
It does not seem too awe-inspiring, but maybe I'm hard to impress.  Or just misinterpreting it. Undecided  
 
Here is a direct link to the article if you can't find it.
http://cipd.mit.edu/documents/publication_files/prelec2004science.pdf
IP Logged

"I do not feel obliged to believe that the same God who has endowed us with sense, reason, and intellect has intended us to forgo their use." - Galileo Galilei
amichail
Senior Riddler
****





   


Posts: 450
Re: Bayesian Truth Serum  
« Reply #2 on: Oct 21st, 2006, 4:42am »
Quote Quote Modify Modify

The task of filling out surveys has been turned into a game where participants try to obtain the highest score possible.
 
The best strategy for winning this game is to tell the truth as best as you can.
 
The whole approach is based on a psychological observation that if you prefer X, then you would overestimate the percentage of people who would also prefer X.
 
The specifics are a bit unclear to me though. Maybe someone could elaborate on why this works, at least in the case where questions have two answers.
« Last Edit: Oct 21st, 2006, 5:22am by amichail » IP Logged

DropZap - a new kind of block elimination game
towr
wu::riddles Moderator
Uberpuzzler
*****



Some people are average, some are just mean.

   


Gender: male
Posts: 13730
Re: Bayesian Truth Serum  
« Reply #3 on: Oct 21st, 2006, 1:07pm »
Quote Quote Modify Modify

on Oct 21st, 2006, 4:42am, amichail wrote:
The whole approach is based on a psychological observation that if you prefer X, then you would overestimate the percentage of people who would also prefer X.
Really?  
I usually assume no one else likes the stuff I like.. But then I've never been that good at playing psychological games Wink
IP Logged

Wikipedia, Google, Mathworld, Integer sequence DB
Grimbal
wu::riddles Moderator
Uberpuzzler
*****






   


Gender: male
Posts: 7527
Re: Bayesian Truth Serum  
« Reply #4 on: Oct 22nd, 2006, 6:36am »
Quote Quote Modify Modify

on Oct 21st, 2006, 1:07pm, towr wrote:
I usually assume no one else likes the stuff I like..

I thought I was the only one like that! Grin
IP Logged
Pages: 1  Reply Reply Notify of replies Notify of replies Send Topic Send Topic Print Print

« Previous topic | Next topic »

Powered by YaBB 1 Gold - SP 1.4!
Forum software copyright © 2000-2004 Yet another Bulletin Board