Reproduce “A Flexible Framework for Incorporating Patient Preferences Into Q-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
Joshua P. Zitovsky, Yating Zou, Leslie Wilson, Michael R. Kosorok
Latent Utility Q-Learning estimates each patient’s latent preferences across competing outcomes from preference reports, then uses those weights in backward optimization to select sequential treatments. In a BEST-style simulation, its value improvement at N=600 was 0.41 (SD 0.09), versus 0.31 and 0.03 for two baselines. This is simulation evidence, not clinical efficacy, and depends on identifiable, well-fitted preference models.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.