homomorpheR: Homomorphic Computations in R
Privacy-preserving statistics across sites that never share
their data, using fully homomorphic encryption through the
'openfhe.R' interface to OpenFHE (CKKS, BFV, BGV), with
n-of-n threshold key generation so that no single party can
decrypt. Ships master/worker primitives that let ordinary R
modeling code run across sites, and a frozen implementation
of the Paillier additive scheme kept for backward
compatibility.
| Version: |
1.0 |
| Depends: |
R (≥ 3.5.0) |
| Imports: |
S7, cli, gmp, openfhe.R, rlang, sodium |
| Suggests: |
knitr, kableExtra, rmarkdown, survival, CVXR, tinytest |
| Published: |
2026-10-03 |
| DOI: |
10.32614/CRAN.package.homomorpheR |
| Author: |
Balasubramanian Narasimhan [aut, cre, cph] |
| Maintainer: |
Balasubramanian Narasimhan <naras at stat.Stanford.EDU> |
| BugReports: |
https://github.com/bnaras/homomorpheR/issues |
| License: |
MIT + file LICENSE |
| URL: |
https://bnaras.github.io/homomorpheR/ |
| NeedsCompilation: |
no |
| Materials: |
README, NEWS |
| CRAN checks: |
homomorpheR results |
Documentation:
| Reference manual: |
homomorpheR.html , homomorpheR.pdf
|
| Vignettes: |
Threshold Cox with Gaussian Noise (Demonstration) (source, R code)
Distributed Cox Regression with Threshold Key Generation (source, R code)
Distributed Stratified Cox Regression (source, R code)
Consensus ADMM with Gaussian Noise (Demonstration) (source, R code)
Federated Cox-Lasso via Consensus ADMM on DLBCL (source, R code)
Encrypted Logistic Regression Prediction (source, R code)
Introduction to homomorpheR (source, R code)
Distributed Maximum Likelihood Estimation (source, R code)
Precision (source, R code)
Privacy-Preserving Count Aggregation (source, R code)
Distributed Query Count with Threshold Keys (source, R code)
Secure Model Inference on Encrypted Data (source, R code)
Federated Cosine Similarity with Site-Private Fine-Tuned Models (source, R code)
|
Downloads:
Reverse dependencies:
Linking:
Please use the canonical form
https://CRAN.R-project.org/package=homomorpheR
to link to this page.