survivalGPU: Scalable Survival Analysis for R and Python
Abstract
This paper introduces survivalGPU, a high-performance Python and R package designed to accelerate survival analysis computations using graphics processing units (GPUs). While the Cox proportional hazards model remains a cornerstone of survival analysis, it can impose substantial computational burdens when applied to large longitudinal data sets. To address this limitation, survivalGPU implements Cox proportional hazards models and weighted cumulative exposure (WCE) models in Python, using PyTorch and KeOps for efficient back-end computation, with interfaces in both Python and R. Numerical validation shows that survivalGPU reproduces the results of reference implementations within the tolerances expected for the selected floating-point precision, with relative errors below 10 -2 in float32 and below 10 -5 in float64. Benchmark experiments demonstrate substantial speedups. For Cox models, coxphGPU() was 5.6× faster than coxph() from the survival package on 1,000,000 patients, and 14.0× faster based on runtimes extrapolated to 1,000 bootstrap replicates on 100,000 patients. Gains were larger for WCE models: wceGPU() was 920.7× faster than WCE() from the WCE package on 100,000 patients, and 1,321.4× faster based on runtimes extrapolated to 1,000 bootstrap replicates. These results show that survivalGPU makes flexible time-dependent survival modeling and bootstrap-based estimation of confidence intervals feasible for large-scale pharmacoepidemiological applications.
Related
- Survival Reinforcement Learning: Toward Scalable Self-Supervised RL
- QuantileMix: A Python package for quantile g-computation
- Exploring continuous beta-ensembles: A Python implementation for random matrix spectral statistics
- OpenTURNS: A Python package for uncertainty quantification
- Eunoia: Area-Proportional Euler and Venn Diagrams in Rust, Julia, Python, R, and JavaScript
BibTeX
@misc{poirotbourdain2026survivalgpu,
title = {{survivalGPU: Scalable Survival Analysis for R and Python}},
author = {Antoine Poirot-Bourdain and Alexis Van Straaten and Emmanuel Bacry and Anne-Sophie Jannot and Jean Feydy},
howpublished = {HAL},
year = {2026},
month = oct,
doi = {10.18637/jss.v000.i00},
url = {https://hal.science/hal-05840926v1},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - survivalGPU: Scalable Survival Analysis for R and Python
AU - Antoine Poirot-Bourdain
AU - Alexis Van Straaten
AU - Emmanuel Bacry
AU - Anne-Sophie Jannot
AU - Jean Feydy
PY - 2026
DA - 2026/10/27
PB - HAL
DO - 10.18637/jss.v000.i00
UR - https://hal.science/hal-05840926v1
AB - This paper introduces survivalGPU, a high-performance Python and R package designed to accelerate survival analysis computations using graphics processing units (GPUs). While the Cox proportional hazards model remains a cornerstone of survival analysis, it can impose substantial computational burdens when applied to large longitudinal data sets. To address this limitation, survivalGPU implements Cox proportional hazards models and weighted cumulative exposure (WCE) models in Python, using PyTorch and KeOps for efficient back-end computation, with interfaces in both Python and R. Numerical validation shows that survivalGPU reproduces the results of reference implementations within the tolerances expected for the selected floating-point precision, with relative errors below 10 -2 in float32 and below 10 -5 in float64. Benchmark experiments demonstrate substantial speedups. For Cox models, coxphGPU() was 5.6× faster than coxph() from the survival package on 1,000,000 patients, and 14.0× faster based on runtimes extrapolated to 1,000 bootstrap replicates on 100,000 patients. Gains were larger for WCE models: wceGPU() was 920.7× faster than WCE() from the WCE package on 100,000 patients, and 1,321.4× faster based on runtimes extrapolated to 1,000 bootstrap replicates. These results show that survivalGPU makes flexible time-dependent survival modeling and bootstrap-based estimation of confidence intervals feasible for large-scale pharmacoepidemiological applications.
ER -