Reproduce “Correlated Quantal Response Equilibrium: A New Equilibrium Notion for Human-AI Strategic Interactions” 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
Zehao Zhao, Rahul Jain, Vijay Subramanian
The authors introduce CQRE-pa, a model for predicting joint actions in strategic games that represents dependence between players and allows limited deviations from recommendations. For each slack setting, it selects a feasible joint distribution with maximum entropy, then fits the slack to data. Across twelve Selten–Chmura human games, KL divergence averages 0.0040 versus 0.0093 for Logit-QRE, indicating better in-sample fit, not out-of-sample prediction.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.