FastBDT: A Speed-Optimized Multivariate Classification Algorithm for the Belle II Experiment
Computing and Software for Big Science · 2017
DOI: 10.1007/s41781-017-0002-8 ↗Kurz gefasst
Fachartikel in der Zeitschrift Computing and Software for Big Science, 2017. Verfasst von Thomas Keck. Die Arbeit wurde bisher 62-mal zitiert (Crossref) und nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Stochastic gradient-boosted decision trees are widely employed for multivariate classification and regression tasks. This paper presents a speed-optimized and cache-friendly implementation for multivariate classification called FastBDT. The concepts used to optimize the execution time are discussed in detail in this paper. The key ideas include: an equal-frequency binning on the input data, which allows replacing expensive floating-point with integer operations, while at the same time increasing the quality of the classification; a cache-friendly linear access pattern to the input data, in contrast to usual implementations, which exhibit a random access pattern. FastBDT provides interfaces to C/C++, Python and TMVA. It is extensively used in the field of high energy physics (HEP) by the Belle II experiment.
Zitieren
APA
Keck, T. (2017). FastBDT: A Speed-Optimized Multivariate Classification Algorithm for the Belle II Experiment. Computing and Software for Big Science. https://doi.org/10.1007/s41781-017-0002-8
Vancouver
Keck T. FastBDT: A Speed-Optimized Multivariate Classification Algorithm for the Belle II Experiment. Computing and Software for Big Science. 2017. doi:10.1007/s41781-017-0002-8
BibTeX
@article{Keck2017,
author = {Keck, Thomas},
title = {FastBDT: A Speed-Optimized Multivariate Classification Algorithm for the Belle II Experiment},
journal = {Computing and Software for Big Science},
year = {2017},
doi = {10.1007/s41781-017-0002-8}
}
EUVIMED