The DARPA SocialSim Challenge: Massive Multi-Agent Simulations of the Github Ecosystem

James Blythe, Emilio Ferrara, Di Huang, Kristina Lerman, Goran Muric, Anna Sapienza, Alexey Tregubov, Diogo Pacheco, John Bollenbacher, Alessandro Flammini, Pik-Mai Hui, Filippo Menczer

Abstract: We model the evolution of GitHub, a large collaborative software-development ecosystem, using massive multi-agent simulations as a part of DARPA's SocialSim program. Our best performing models and our agent-based simulation framework are described here. Six different agent models were tested based on a variety of machine learning and statistical methods. The most successful models are based on sampling from a stationary probability distribution of actions and repositories for each agent.