Reproduce “Manipulation-Proof Machine 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
Daniel Björkegren, Joshua E. Blumenstock, Samsun Knight
The authors develop a strategy-robust model that predicts outcomes after people respond to a decision rule. It estimates the costs of changing behavior from randomized incentives, then selects a rule for the behavior people are likely to exhibit. In a Kenyan smartphone-app experiment, transparent implementation yielded pooled RMSE of $4.31 versus $4.93 for LASSO; evidence is limited to this setting.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.