Reproduce “First-Order Bilevel Optimization for Non-truthful Auction Design” 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
Shing-Hei Ho, April Niu, Ellen Vitercik, Kai Wang
The authors optimize pay-your-bid auctions while accounting for how bidders’ equilibrium strategies respond to changes in the mechanism. Their first-order method estimates this strategic response using an auxiliary perturbed game, avoiding explicit second-order information but requiring an additional game solve. In a symmetric two-bidder, one-item Uniform[0,1] setting, revenue is 0.404, close to 0.417 for the optimal truthful auction.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.