Show simple item record  

dc.contributor.authorMitchell, Roryen_NZ
dc.contributor.authorFrank, Eibeen_NZ
dc.contributor.authorHolmes, Geoffreyen_NZ
dc.date.accessioned2022-04-07T02:25:46Z
dc.date.available2022-04-07T02:25:46Z
dc.date.issued2022en_NZ
dc.identifier.issn2376-5992en_NZ
dc.identifier.urihttps://hdl.handle.net/10289/14815
dc.description.abstractSHapley Additive exPlanation (SHAP) values (Lundberg & Lee, 2017) provide a game theoretic interpretation of the predictions of machine learning models based on Shapley values (Shapley, 1953). While exact calculation of SHAP values is computationally intractable in general, a recursive polynomial-time algorithm called TreeShap (Lundberg et al., 2020) is available for decision tree models. However, despite its polynomial time complexity, TreeShap can become a significant bottleneck in practical machine learning pipelines when applied to large decision tree ensembles. Unfortunately, the complicated TreeShap algorithm is difficult to map to hardware accelerators such as GPUs. In this work, we present GPUTreeShap, a reformulated TreeShap algorithm suitable for massively parallel computation on graphics processing units. Our approach first preprocesses each decision tree to isolate variable sized sub-problems from the original recursive algorithm, then solves a bin packing problem, and finally maps sub-problems to single-instruction, multiple-thread (SIMT) tasks for parallel execution with specialised hardware instructions. With a single NVIDIA Tesla V100-32 GPU, we achieve speedups of up to 19× for SHAP values, and speedups of up to 340× for SHAP interaction values, over a state-of-the-art multi-core CPU implementation executed on two 20-core Xeon E5-2698 v4 2.2 GHz CPUs. We also experiment with multi-GPU computing using eight V100 GPUs, demonstrating throughput of 1.2 M rows per second—equivalent CPU-based performance is estimated to require 6850 CPU cores.
dc.format.mimetypeapplication/pdf
dc.language.isoen
dc.publisherPeerJen_NZ
dc.rights© Copyright 2022 Mitchell et al. This article is published under the Creative Commons CC-BY 4.0
dc.subjectGPU computingen_NZ
dc.subjectShapley valuesen_NZ
dc.subjectinterpretabilityen_NZ
dc.subjectcomputer scienceen_NZ
dc.titleGPUTreeShap: massively parallel exact calculation of SHAP scores for tree ensemblesen_NZ
dc.typeJournal Article
dc.identifier.doi10.7717/peerj-cs.880en_NZ
dc.relation.isPartOfPeerJ Computer Scienceen_NZ
pubs.begin-pagee880
pubs.elements-id262233
pubs.end-pagee880
pubs.publication-statusPublisheden_NZ
pubs.volume8en_NZ


Files in this item

This item appears in the following Collection(s)

Show simple item record