A survey on FPGA-based accelerator for ML models

TL;DR

This survey reviews recent FPGA-based hardware accelerators for machine learning, highlighting trends, dominant models like CNN, and emerging areas such as GNN, based on an analysis of 287 selected papers from top FPGA conferences.

Contribution

It provides a comprehensive overview of FPGA-based ML acceleration research, categorizing topics, identifying trends, and highlighting the dominance of inference acceleration and CNN models.

Findings

Inference acceleration dominates (81%)

CNN is the most researched model

Emerging interest in GNNs

Abstract

This paper thoroughly surveys machine learning (ML) algorithms acceleration in hardware accelerators, focusing on Field-Programmable Gate Arrays (FPGAs). It reviews 287 out of 1138 papers from the past six years, sourced from four top FPGA conferences. Such selection underscores the increasing integration of ML and FPGA technologies and their mutual importance in technological advancement. Research clearly emphasises inference acceleration (81\%) compared to training acceleration (13\%). Additionally, the findings reveals that CNN dominates current FPGA acceleration research while emerging models like GNN show obvious growth trends. The categorization of the FPGA research papers reveals a wide range of topics, demonstrating the growing relevance of ML in FPGA research. This comprehensive analysis provides valuable insights into the current trends and future directions of FPGA research in the context of ML applications.