On-sensor Printed Machine Learning Classification via Bespoke ADC and Decision Tree Co-Design

Giorgos Armeniakos, Paula L. Duarte, Priyanjana Pal, Georgios Zervakis, Mehdi B. Tahoori +1 more
12/2/2023
cs.LG

Official code: garmeniakos/Ax-Printed-ML-Classifiers (7 stars, last commit 2024-04-15).

Abstract

Printed electronics (PE) technology provides cost-effective hardware with unmet customization, due to their low non-recurring engineering and fabrication costs. PE exhibit features such as flexibility, stretchability, porosity, and conformality, which make them a prominent candidate for enabling ubiquitous computing. Still, the large feature sizes in PE limit the realization of complex printed circuits, such as machine learning classifiers, especially when processing sensor inputs is necessary, mainly due to the costly analog-to-digital converters (ADCs). To this end, we propose the design of fully customized ADCs and present, for the first time, a co-design framework for generating bespoke Decision Tree classifiers. Our comprehensive evaluation shows that our co-design enables self-powered operation of on-sensor printed classifiers in all benchmark cases.

View on arXivView PDF

Code Implementations(1)

garmeniakos/Ax-Printed-ML-ClassifiersOfficial100%

Approximate Printed Machine Learning Classifiers

72Python, VerilogDec 13, 20212 years ago
approximate-computingcross-layercross-layer-approximationdesign-automationdesign-space-exploration+2 more

Similar Papers

Model-to-Circuit Cross-Approximation For Printed Machine Learning Classifiers

83% match2023

Hybrid unary-binary design for multiplier-less printed Machine Learning classifiers

82% match2025

Compact Yet Highly Accurate Printed Classifiers Using Sequential Support Vector Machine Circuits

82% match2025

Co-Design of Approximate Multilayer Perceptron for Ultra-Resource Constrained Printed Circuits

81% match1 repo72023

Late Breaking Results: Energy-Efficient Printed Machine Learning Classifiers with Sequential SVMs

80% match2025

Bespoke Approximation of Multiplication-Accumulation and Activation Targeting Printed Multilayer Perceptrons

77% match1 repo2023

Cite this paper

@article{armeniakos2023onsensor,
  title  = {On-sensor Printed Machine Learning Classification via Bespoke ADC and Decision Tree Co-Design},
  author = {Giorgos Armeniakos and Paula L. Duarte and Priyanjana Pal and Georgios Zervakis and Mehdi B. Tahoori and Dimitrios Soudris},
  year   = {2023},
  eprint = {2312.01172},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2312.01172v1}
}

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/2312.01172.svg)](https://codeofpaper.com/paper/2312.01172)

Discussion