Evolutionary Approximation of Ternary Neurons for On-sensor Printed Neural Networks
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911971523166208 |
|---|---|
| author | Mrazek, Vojtech Kokkinis, Argyris Papanikolaou, Panagiotis Vasicek, Zdenek Siozios, Kostas Tzimpragos, Georgios Tahoori, Mehdi Zervakis, Georgios |
| author_facet | Mrazek, Vojtech Kokkinis, Argyris Papanikolaou, Panagiotis Vasicek, Zdenek Siozios, Kostas Tzimpragos, Georgios Tahoori, Mehdi Zervakis, Georgios |
| contents | Printed electronics offer ultra-low manufacturing costs and the potential for on-demand fabrication of flexible hardware. However, significant intrinsic constraints stemming from their large feature sizes and low integration density pose design challenges that hinder their practicality. In this work, we conduct a holistic exploration of printed neural network accelerators, starting from the analog-to-digital interface - a major area and power sink for sensor processing applications - and extending to networks of ternary neurons and their implementation. We propose bespoke ternary neural networks using approximate popcount and popcount-compare units, developed through a multi-phase evolutionary optimization approach and interfaced with sensors via customizable analog-to-binary converters. Our evaluation results show that the presented designs outperform the state of the art, achieving at least 6x improvement in area and 19x in power. To our knowledge, they represent the first open-source digital printed neural network classifiers capable of operating with existing printed energy harvesters. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_20589 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Evolutionary Approximation of Ternary Neurons for On-sensor Printed Neural Networks Mrazek, Vojtech Kokkinis, Argyris Papanikolaou, Panagiotis Vasicek, Zdenek Siozios, Kostas Tzimpragos, Georgios Tahoori, Mehdi Zervakis, Georgios Hardware Architecture Printed electronics offer ultra-low manufacturing costs and the potential for on-demand fabrication of flexible hardware. However, significant intrinsic constraints stemming from their large feature sizes and low integration density pose design challenges that hinder their practicality. In this work, we conduct a holistic exploration of printed neural network accelerators, starting from the analog-to-digital interface - a major area and power sink for sensor processing applications - and extending to networks of ternary neurons and their implementation. We propose bespoke ternary neural networks using approximate popcount and popcount-compare units, developed through a multi-phase evolutionary optimization approach and interfaced with sensors via customizable analog-to-binary converters. Our evaluation results show that the presented designs outperform the state of the art, achieving at least 6x improvement in area and 19x in power. To our knowledge, they represent the first open-source digital printed neural network classifiers capable of operating with existing printed energy harvesters. |
| title | Evolutionary Approximation of Ternary Neurons for On-sensor Printed Neural Networks |
| topic | Hardware Architecture |
| url | https://arxiv.org/abs/2407.20589 |