Skip to content

kunanit/mcmc-with-colors

Repository files navigation

mcmc-with-colors

Determining subjective beliefs on color distributions by using people as elements of a MCMC algorithm.

Created as a project for Harvard CS 186 (Economics and Computation), Spring 2015.

Report

View the report

Simulation and data analysis

View ipython notebooks on nbviewer:

Web application

Try out the color choice task at mcmc-with-colors.herokuapp.com. The web application is built using Django, a python web framework.

References:

Adam Sanborn and Thomas Griffiths, "Markov Chain Monte Carlo with People" (NIPS 2008)

About

Human computation of MCMC via a color picking task

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published