Skip to main navigation Skip to search Skip to main content

HiFi-SAGE: High Fidelity GraphSAGE-Based Latency Estimators for DNN Optimization

  • Shambhavi Balamuthu Sampath
  • , Leon Hecht
  • , Moritz Thoma
  • , Lukas Frickenstein
  • , Pierpaolo Mori
  • , Nael Fasfous
  • , Manoj Rohit Vemparala
  • , Alexander Frickenstein
  • , Walter Stechele
  • , Daniel Mueller-Gritschneder
  • , Claudio Passerone
  • Technical University of Munich
  • Innovations
  • Politecnico di Torino

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

As deep neural networks (DNNs) are increasingly deployed on resource-constrained edge devices, optimizing and compressing them for real-time performance becomes crucial. Traditional hardware-aware DNN search methods often rely on inaccurate proxy metrics, expensive latency lookup tables, or slow hardware-in-the-Iloop (HIL) evaluations. To address this, quasi-generalized latency estimators, typically meta-learning-based, were proposed to replace HIL evaluations and accelerate the search. These come with a one-time data collection and training cost and can adapt to new hardware with few measurements. However, they still have some drawbacks: (1) They increase complexity by trying to generalize across a range of diverse hardware types; (2) They depend on handcrafted hardware descriptors, which may fail to capture hardware characteristics; (3) They often perform poorly on new, unseen hardware that significantly differs from their initial training set. To overcome these challenges, this paper turns to the more straightforward platform-specific estimators that do not require hardware descriptors and can be easily trained on any hardware. We introduce HiFi-SAGE, a high fidelity GraphSAGE-based platform-specific latency estimator. When trained from scratch on only 100 latency measurements, our novel dual-head estimator design surpasses the state-of-the-art (SoTA) on the 10% error bound metric by up to 17.4 p.p. while achieving an impressive fidelity score of 99% on the diverse LatBench dataset. We demonstrate that applying HiFi-SAGE to a genetic algorithm-based DNN compression search, achieved a Pareto front comparable to real HIL feedback with a mean absolute percentage error (MAPE) of 2.54%, 2.48%, and 4.16%, for InceptionV3, DenseNet169, and ResNet50 respectively. Compared to existing platform-specific works, the lower number of latency measurements and higher fidelity scores positions HiFi-SAGE as an attractive alternative to replace expensive HIL setups. Code is available at: https://github.com/shamvbs/HiFi-SAGE

Original languageEnglish
Title of host publication2025 Design, Automation and Test in Europe Conference, DATE 2025 - Proceedings
PublisherInstitute of Electrical and Electronics Engineers Inc.
ISBN (Electronic)9783982674100
DOIs
StatePublished - 2025
Event2025 Design, Automation and Test in Europe Conference, DATE 2025 - Lyon, France
Duration: 31 Mar 20252 Apr 2025

Publication series

NameProceedings -Design, Automation and Test in Europe, DATE
ISSN (Print)1530-1591

Conference

Conference2025 Design, Automation and Test in Europe Conference, DATE 2025
Country/TerritoryFrance
CityLyon
Period31/03/252/04/25

Fingerprint

Dive into the research topics of 'HiFi-SAGE: High Fidelity GraphSAGE-Based Latency Estimators for DNN Optimization'. Together they form a unique fingerprint.

Cite this