This repository contains all the files necessary to reproduce the paper Expected Values Estimated via Mean-Field Approximation are 1/N-Accurate that has been accepted for at ACM SIGMETRICS conference 2017.
Full reference : Nicolas Gast. Expected Values Estimated via Mean-Field Approximation are 1/N-Accurate. Proceedings of ACM SIGMETRICS conference 2017. Urbana-Champaign, USA, June 6-8, 2017 (SIGMETRICS’17).
Publisher version : https://github.com/ngast/meanFieldAccuracy/releases/tag/finalVersion
How to compile
To compile the paper:
The directories jsq2_simulate and simu contains the files necessary to generate the figures.
- jsq2_simulate contains:
- a file "two_choice_simulator.cc" that is a C++ implementation of a simulator of the two-choice model
- an "ipython"-notebook that is used to generate simulation data and the figures
- simu contains the files to generates the figures of the small 1 dimensional example of Section 4.
To compile the simulator and re-generate all figures, just type
make simulator make simulations
Note that the directory
jsq2_simulate/traj contain simulation results that are not re-generated each time (in order to save time). Removing these files and launching
make simulations (or executing directely the "jupyter notebook"
jsq2_simulate/Rate of convergence two choice.ipynb) will regenerate these files. This might take a while (~10h).
An extended abstract of the paper is also available.