arrow
Return

Accelerating GridSearchCV hyperparameter tuning method using an FPGA-based hardware accelerator

delete2025-12-06
delete0
PRE
AI
A
Arian Shahbazian
M
Mohammadreza Binesh Marvasti *
S
Seyyed Amir Asghari
DOI:10.1007/s11227-025-08124-7delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
Machine learning models underpin critical applications, and their performance hinges on effective hyperparameter tuning, yet exhaustive methods like GridSearchCV are computationally intensive on general-purpose systems. This paper presents a custom FPGA accelerator for the GridSearchCV–KNN pipeline, simulated on an Intel Cyclone V platform. The six-module RTL design features an on-chip unique memory unit, a pipelined Manhattan distance subtractor, and a parallel sorting Distance Memory that tracks nearest neighbors without explicit sorting. A dedicated hardware controller manages hyperparameter grid enumeration and K-fold evaluation entirely in hardware. We introduce ’CV-full’ (Leave-One-Out Cross-Validation) to test our system under the most intensive circumstances. Post-synthesis simulations show average speedups of 1.8 $$\times$$ , 2 $$\times$$ , and about 10 $$\times$$ under 5-fold, 10-fold, and CV-full settings, respectively, with a peak of 27.7 $$\times$$ over a Google Colab server. By minimizing off-chip memory access and explicit sorting stages, our FPGA design offers a special-purpose hardware solution for accelerating GridSearchCV–KNN hyperparameter tuning.
Keywords:
FPGA
GridSearchCV
Hardware accelerator
Hyperparameter tuning
Machine learning

Journal

T
The Journal of Supercomputing
IF:
0
Papers:
647
Citations:
0

Organization

D
Department of Electrical and Computer Engineering
Scholars:
812
Papers: 433
Citations: 6
D
data science and technology department
Scholars:
1
Papers: 1
Citations: 0
Cited Papers

Cited Papers

Hardware Acceleration for Object Detection using YOLOv5 Deep Learning Algorithm on Xilinx Zynq FPGA Platform
err2024-02-08
err0
errOAAI
errTaoufik Saidani; Refka Ghodhbani; Ahmed Alhomoud; Ahmad Alshammari; Hafedh Zayani; Mohammed Ben Ammar
errShare
errSave
err
IF0
err
err0
PREAI
err
errShare
errSave
researcher View more