An Open-Source ML-Based Full-Stack Optimization Framework for Machine Learning Accelerators

Hadi Esmaeilzadeh, Soroush Ghodrati, Andrew B. Kahng, Joon Kyung Kim, Sean Kinzer +6 more
8/23/2023
cs.LGcs.AR

No official code found. Best implementation: VeriGOOD-ML/public (71 stars, Apache-2.0, last commit 2025-04-30).

Abstract

Parameterizable machine learning (ML) accelerators are the product of recent breakthroughs in ML. To fully enable their design space exploration (DSE), we propose a physical-design-driven, learning-based prediction framework for hardware-accelerated deep neural network (DNN) and non-DNN ML algorithms. It adopts a unified approach that combines backend power, performance, and area (PPA) analysis with frontend performance simulation, thereby achieving a realistic estimation of both backend PPA and system metrics such as runtime and energy. In addition, our framework includes a fully automated DSE technique, which optimizes backend and system metrics through an automated search of architectural and backend parameters. Experimental studies show that our approach consistently predicts backend PPA and system metrics with an average 7% or less prediction error for the ASIC implementation of two deep learning accelerator platforms, VTA and VeriGOOD-ML, in both a commercial 12 nm process and a research-oriented 45 nm process.

View on arXivView PDF

Code Implementations(1)

VeriGOOD-ML/publicCommunity100%
Apache-2.0
7113Tcl, ShellApr 4, 20211 years agoApache-2.0

Similar Papers

A Survey on Design Methodologies for Accelerating Deep Learning on Heterogeneous Architectures

89% match2023

FPGA-based Acceleration for Convolutional Neural Networks: A Comprehensive Review

88% match2025

GANDSE: Generative Adversarial Network based Design Space Exploration for Neural Network Accelerator Design

86% match2022

MetaML-Pro: Cross-Stage Design Flow Automation for Efficient Deep Learning Acceleration

85% match2025

Exploration of TPUs for AI Applications

85% match2023

Machine Learning aided Computer Architecture Design for CNN Inferencing Systems

85% match2023

Cite this paper

@article{esmaeilzadeh2023opensource,
  title  = {An Open-Source ML-Based Full-Stack Optimization Framework for Machine Learning Accelerators},
  author = {Hadi Esmaeilzadeh and Soroush Ghodrati and Andrew B. Kahng and Joon Kyung Kim and Sean Kinzer and Sayak Kundu and Rohan Mahapatra and Susmita Dey Manasi and Sachin Sapatnekar and Zhiang Wang and Ziqing Zeng},
  year   = {2023},
  eprint = {2308.12120},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2308.12120v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2308.12120.svg)](https://codeofpaper.com/paper/2308.12120)

Discussion