Unknown

Dataset Information

0

Bayesian Lasso and multinomial logistic regression on GPU.


ABSTRACT: We describe an efficient Bayesian parallel GPU implementation of two classic statistical models-the Lasso and multinomial logistic regression. We focus on parallelizing the key components: matrix multiplication, matrix inversion, and sampling from the full conditionals. Our GPU implementations of Bayesian Lasso and multinomial logistic regression achieve 100-fold speedups on mid-level and high-end GPUs. Substantial speedups of 25 fold can also be achieved on older and lower end GPUs. Samplers are implemented in OpenCL and can be used on any type of GPU and other types of computational units, thereby being convenient and advantageous in practice compared to related work.

SUBMITTER: Cesnovar R 

PROVIDER: S-EPMC5489220 | biostudies-literature | 2017

REPOSITORIES: biostudies-literature

altmetric image

Publications

Bayesian Lasso and multinomial logistic regression on GPU.

Češnovar Rok R   Štrumbelj Erik E  

PloS one 20170628 6


We describe an efficient Bayesian parallel GPU implementation of two classic statistical models-the Lasso and multinomial logistic regression. We focus on parallelizing the key components: matrix multiplication, matrix inversion, and sampling from the full conditionals. Our GPU implementations of Bayesian Lasso and multinomial logistic regression achieve 100-fold speedups on mid-level and high-end GPUs. Substantial speedups of 25 fold can also be achieved on older and lower end GPUs. Samplers ar  ...[more]

Similar Datasets

| S-EPMC3771412 | biostudies-literature
| S-EPMC4908321 | biostudies-other
| S-EPMC2732298 | biostudies-literature
| S-EPMC4174302 | biostudies-literature
| S-EPMC9256608 | biostudies-literature
| S-EPMC4144467 | biostudies-literature
| S-EPMC6329526 | biostudies-literature
| S-EPMC6156907 | biostudies-literature
| S-EPMC6175404 | biostudies-literature
| S-EPMC6836722 | biostudies-literature