Reproduce “Deconstructing Cooperation and Ostracism via Multi-Agent Reinforcement Learning” 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

Atsushi Ueshima, Shayegan Omidshafiei, Hirokazu Shirado

The paper tests whether simulated agents can sustain cooperation by learning interaction and rewiring policies in a Prisoner’s Dilemma. Rewiring occurs first, and links require both agents to connect. With full rewiring, learned policies let TFT-biased agents reach 100% mutual cooperation within 30,000 episodes; disabling rewiring-policy learning leaves cooperation almost absent. This supports learned rewiring’s importance in symmetric two-agent simulations, not ostracism alone.

Summary by alphaXiv.

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