Reproduce “Self-Resolving Prediction Markets for Unverifiable Outcomes” 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

Siddarth Srinivasan, Ezra Karger, Yiling Chen

This paper presents a mechanism for self-resolving prediction markets that can operate and incentivize truthful information aggregation even when the predicted outcome is unverifiable. The proposed system achieves incentive compatibility by leveraging the information from subsequent agents to form a robust internal reference for payment, expanding the scope of prediction markets to domains like policy evaluation and long-term forecasting.

Key insights

  • A sufficiently informed reference agent can serve as a robust proxy for ground truth, enabling payout resolution without external verification.
  • Strategic influence by an agent on the reference agent's belief diminishes exponentially as the reference agent observes more independent informational substitutes.
  • Cross-entropy market scoring rules can be designed to make uninformative reporting strategies yield zero payoff, thus incentivizing truthful, informative participation.

Summary by alphaXiv.

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