Reproduce “Designing Experiments for External Decision Makers” 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
Sahel Iqbal, Tom Rainforth
The authors introduce usage-aware Bayesian experimental design, which optimizes data collection for a fixed downstream decision rule rather than the experimenter’s preferred action. It recovers standard design when the rule is the experimenter’s Bayes act. In a simulated trial, adaptive design yields €57.79 million versus €52.27 million for the matched single-stage design; this is not a real-trial outcome.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.