FPGA accelerator for gradient boosting decision trees

Alcolea, A. (Universidad de Zaragoza) ; Resano, J. (Universidad de Zaragoza)

Resumen: A decision tree is a well-known machine learning technique. Recently their popularity has increased due to the powerful Gradient Boosting ensemble method that allows to gradually increasing accuracy at the cost of executing a large number of decision trees. In this paper we present an accelerator designed to optimize the execution of these trees while reducing the energy consumption. We have implemented it in an FPGA for embedded systems, and we have tested it with a relevant case-study: pixel classification of hyperspectral images. In our experiments with different images our accelerator can process the hyperspectral images at the same speed at which they are generated by the hyperspectral sensors. Compared to a high-performance processor running optimized software, on average our design is twice as fast and consumes 72 times less energy. Compared to an embedded processor, it is 30 times faster and consumes 23 times less energy.

Idioma: Inglés

DOI: 10.3390/electronics10030314

Año: 2021

Publicado en: Electronics 10, 3 (2021), 314 [15 pp]
ISSN: 2079-9292

Factor impacto JCR: 2.69 (2021)

Categ. JCR: COMPUTER SCIENCE, INFORMATION SYSTEMS rank: 100 / 164 = 0.61 (2021) - Q3 - T2
Categ. JCR: PHYSICS, APPLIED rank: 82 / 161 = 0.509 (2021) - Q3 - T2
Categ. JCR: ENGINEERING, ELECTRICAL & ELECTRONIC rank: 139 / 277 = 0.502 (2021) - Q3 - T2

Factor impacto CITESCORE: 3.7 - Computer Science (Q2) - Engineering (Q2)

Factor impacto SCIMAGO: 0.59 - Computer Networks and Communications (Q2) - Signal Processing (Q2) - Hardware and Architecture (Q2) - Control and Systems Engineering (Q2)

Financiación: info:eu-repo/grantAgreement/ES/AEI-FEDER/PID2019-105660RB-C21
Financiación: info:eu-repo/grantAgreement/ES/AEI-FEDER/TIN2016-76635-C2-1-R
Financiación: info:eu-repo/grantAgreement/ES/DGA-ESF/T58-20R
Financiación: info:eu-repo/grantAgreement/ES/DGA-FEDER/Construyendo Europa desde Aragón

Tipo y forma: Article (Published version)

Área (Departamento): Área Arquit.Tecnología Comput. (Dpto. Informát.Ingenie.Sistms.)