Stanford AI for Human Impact

`kpe` — K-Fold Personalization Estimator

Confidence intervals and a hypothesis test for the value of personalizing treatment decisions.

Emma Brunskill & Zhaoqi Li · Stanford University

Packages · install from GitHub

kpe-pyPython Docs GitHub

pip install "git+https://github.com/StanfordAI4HI/kpe-py.git"  # then: from kpe import kpe

kpe-rR Docs GitHub

pak::pak("StanfordAI4HI/kpe-r")  # then: library(kpe)

Both install from GitHub. Not yet on PyPI / CRAN — planned. (The importable package is kpe in both languages.)

Learn more

Zhaoqi Li, Emma Brunskill, A statistical test for the benefits of personalizing interventions. Science 393, eaeb9506 (2026). DOI: 10.1126/science.aeb9506
2-fold variant detailed here

License

kpe-py and kpe-r are both released under the MIT License — provided “as is”, without warranty of any kind; the authors are not liable for any claim or damages.