LGBM2VHDL: Mapping of LightGBM Models to FPGA

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F00216305%3A26230%2F24%3APU154643" target="_blank" >RIV/00216305:26230/24:PU154643 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/63839172:_____/24:10133691

Výsledek na webu

<a href="http://dx.doi.org/10.1109/FCCM60383.2024.00020" target="_blank" >http://dx.doi.org/10.1109/FCCM60383.2024.00020</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.1109/FCCM60383.2024.00020" target="_blank" >10.1109/FCCM60383.2024.00020</a>

Jazyk výsledku

angličtina

Název v původním jazyce

LGBM2VHDL: Mapping of LightGBM Models to FPGA

Popis výsledku v původním jazyce

Gradient boosting (GB) is an effective and widely used type of ensemble machine-learning method. The opportunity to transform the trained GB models to the hardware level represents the potential for significant acceleration of many applications and their availability as embedded systems. In this work, we have therefore developed the LGBM2VHDL tool for the automated mapping of models trained by the LightGBM library to circuits described by VHDL. Compared to existing tools, we have used an architecture that is better suited for large-scale GB models involving up to thousands of decision trees. We have further optimized the architecture using two newly proposed techniques. By applying these techniques to the tested models, the amount of memory required was significantly reduced to almost half of the original resources, and the amount of basic configurable blocks was reduced by up to 4 times on average. The developed tool is available as open-source.

Název v anglickém jazyce

LGBM2VHDL: Mapping of LightGBM Models to FPGA

Popis výsledku anglicky

Gradient boosting (GB) is an effective and widely used type of ensemble machine-learning method. The opportunity to transform the trained GB models to the hardware level represents the potential for significant acceleration of many applications and their availability as embedded systems. In this work, we have therefore developed the LGBM2VHDL tool for the automated mapping of models trained by the LightGBM library to circuits described by VHDL. Compared to existing tools, we have used an architecture that is better suited for large-scale GB models involving up to thousands of decision trees. We have further optimized the architecture using two newly proposed techniques. By applying these techniques to the tested models, the amount of memory required was significantly reduced to almost half of the original resources, and the amount of basic configurable blocks was reduced by up to 4 times on average. The developed tool is available as open-source.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20206 - Computer hardware and architecture

Projekt

<a href="/cs/project/VJ02010024" target="_blank" >VJ02010024: Analýza šifrovaného provozu pomocí síťových toků</a><br>

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2024

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

2024 IEEE 32nd Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)

ISBN

979-8-3503-7243-4

ISSN

—

e-ISSN

—

Počet stran výsledku

7

Strana od-do

97-103

Název nakladatele

IEEE Computer Society

Místo vydání

Orlando, FL

Místo konání akce

Orlando, FL

Datum konání akce

5. 5. 2024

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

—