Accelerating the XGBoost algorithm using GPU computing | Zendy

Rory Mitchell | Zendy; Eibe Frank | Zendy

AI Assistant Blog Pricing

Home ZAIA Blog

Open Access

Accelerating the XGBoost algorithm using GPU computing

Author(s) -

Rory Mitchell,

Eibe Frank

Publication year - 2017

Publication title -

peerj computer science

Language(s) - English

Resource type - Journals

SCImago Journal Rank - 0.806

H-Index - 24

ISSN - 2376-5992

DOI - 10.7717/peerj-cs.127

Subject(s) - cuda , computer science , xeon , titan (rocket family) , parallel computing , graphics processing unit , boosting (machine learning) , general purpose computing on graphics processing units , xeon phi , algorithm , sort , speedup , decision tree , gradient boosting , multi core processor , graphics , artificial intelligence , computer graphics (images) , random forest , database , engineering , aerospace engineering

We present a CUDA-based implementation of a decision tree construction algorithm within the gradient boosting library XGBoost. The tree construction algorithm is executed entirely on the graphics processing unit (GPU) and shows high performance with a variety of datasets and settings, including sparse input matrices. Individual boosting iterations are parallelised, combining two approaches. An interleaved approach is used for shallow trees, switching to a more conventional radix sort-based approach for larger depths. We show speedups of between 3x and 6x using a Titan X compared to a 4 core i7 CPU, and 1.2x using a Titan X compared to 2x Xeon CPUs (24 cores). We show that it is possible to process the Higgs dataset (10 million instances, 28 features) entirely within GPU memory. The algorithm is made available as a plug-in within the XGBoost library and fully supports all XGBoost features including classification, regression and ranking tasks

The content you want is available to Zendy users.

Already have an account? Click here to sign in.

Having issues? You can contact us here

Accelerating Research