Reproduce “Yielding Fairness: Coupled Learning and Fairness for Value-Weighted Ranking” 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
Ami Tavory, Yaniv Sheena, Shaked Bar, Yair Zagdansky
The authors present a ranking method for blocking accounts that balances expected loss prevented against value forgone. It penalizes blocks that a specified alternative criterion would pass at the same value budget, with penalties adjustable by group. At 95% validation capture floor, group-aware pricing had a ten-source mean closure lift +0.0863, but no statistically significant average advantage over group-blind pricing.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.