Grace Guan

Grace Guan

PhD Student at Stanford University

Research

I am interested in AI safety and societal impacts.

Organ Allocation

My PhD research was on matching and market design using data science tools (machine learning, experimentation). I worked with Itai Ashlagi and Nobel laureate Al Roth on improving the complex deceased donor kidney allocation system and had real-world policy impact.

Specifically, I developed heuristics [1] and a machine learning tool [2, 3] to predict when a deceased donor kidney would be at risk of nonuse. I designed a switchback experiment [4] with organ procurement organizations to assess the causal effect of the machine learning tool on medical decision making. I also designed alternative allocation policies with principles from mechanism design to improve efficiency and equity [5]. I built a kidney allocation policy simulator in C/C++/Python to evaluate these alternative policies.

A subset of publications from this work:

  1. Guan et al. Insights from refusal patterns for deceased donor kidney offers. Transplantation, 2025. 💡 Market deisgn blog link 💡

  2. Guan, et al. Machine learning predictions for assessing hard-to-place deceased donor kidneys. Kidney Medicine, 2025.

  3. Agarwal, Ashlagi, Guan, Somaini, Zou. Time-constrained decision making in deceased donor kidney allocation 🌟 NeurIPS 2022 Workshop on Learning from Time Series for Health (spotlight presentation) 🌟

  4. Agarwal, Ashlagi, Guan, Somaini, Zou. Artificial Intelligence to Enhance Organ Procurement Organization Coordinators’ Decisions to Expedite. Technical Report.

  5. Guan, et al. Accelerated allocation policies to improve utilization of marginal deceased donor kidneys In preparation.

Other selected research

I have also worked on various projects in hospital operations and health policy. For a full list of my publications, please visit my Google Scholar.

Empirical characteristics of Affordable Care Act risk transfers
Grace Guan, advised by Mark Braverman
Princeton Computer Science Senior Thesis
🏆 Won Outstanding Computer Science Senior Thesis Prize
🏆 Won Princeton’s Choice Award at Princeton Research Day 2018

Guan and Engelhardt. Predicting sick patient volume in a pediatric outpatient setting using time series analysis. MLHC 2019: Proceedings of Machine Learning Research, 2019, 106: 271-287
🏆 Won 2020 NCWIT Collegiate Award Honorable Mention