Reproduce “Payoff-Aware Prediction of Population Game Dynamics” on OpenResearch

Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.

Universal app for Apple silicon and Intel. Requires macOS 11 or later.

About this paper

Kang Wang, Xiao Wang, Bo Li, Renzhe Xu

PA-RmD forecasts population trajectories under unseen payoff functions by combining an explicit payoff signal with a learned response structure shared across payoff settings. In a four-strategy zero-sum cyclic game, its held-out full-trajectory rollout RMSE is 0.0173 ± 0.0009, versus 0.0934 ± 0.0099 for zero-shot CoDA-ℓ₂. Transfer depends on shared responses and payoff coverage; error rises when responses vary with payoffs.

Summary by alphaXiv.

Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.