In the DARPA SocialSim GitHub challenge, per-user stationary probabilistic agents broadly outperformed graph-embedding and Bayesian agents, because GitHub users' behavior changes slowly.
In: 5th International Symposium on Resilient Control Systems
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.MA 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Massive Multi-Agent Data-Driven Simulations of the GitHub Ecosystem
In the DARPA SocialSim GitHub challenge, per-user stationary probabilistic agents broadly outperformed graph-embedding and Bayesian agents, because GitHub users' behavior changes slowly.