Neural Evolutionary Architecture Search for Compact Printed Analog Neuromorphic Circuits
Uložené v:
| Názov: | Neural Evolutionary Architecture Search for Compact Printed Analog Neuromorphic Circuits |
|---|---|
| Autori: | Haibin Zhao, Priyanjana Pal, Michael Hefenbrock, Yuhong Wang, Michael Beigl, Mehdi B. Tahoori |
| Zdroj: | IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. 44:2655-2668 |
| Informácie o vydavateľovi: | Institute of Electrical and Electronics Engineers (IEEE), 2025. |
| Rok vydania: | 2025 |
| Predmety: | ddc:004, neural architecture search, compact circuit design, machine learning, evolutionary algorithm, gradient-based, DATA processing & computer science, printed electronics, optimization, neuromorphic computing |
| Popis: | Printed electronics (PE) is an additive fabrication technology for manufacturing electronic circuits which not only allows for a highly flexible printing of arbitrary circuit patterns, but also produce soft, non-toxic, and degradable electronics at an extremely low cost. These properties are unmatched by siliconbased electronics, making PE an enabler of new application domains, e.g., fast moving consumer goods, wearables, and disposable healthcare devices. A particularly promising class of circuits in this technology is the printed analog neuromorphic circuits, offering efficient and highly tailored computational functionalities. In this work, we leverage the highly flexible fabrication process of PE to address the bottleneck of PE, i.e., the large feature sizes and low device counts. This issue is crucial, as it impairs the integration of printed circuits into target applications with limited footprint, such as smart bandaids. We also propose an evolutionary algorithm (EA) to improve the circuit compactness through circuit architecture optimization. As baseline, we compare the proposed EA method with a stateof-the-art pruning method and a modified area-aware pruning method. All of them are able to optimize circuit architecture alongside the component values of printed neuromorphic circuits. Experimental simulation reveals that the proposed EA approach can effectively achieve compact circuits and outperform the pruning method by 3.1× lower area with no loss of accuracy. As a byproduct, the power is reduced by 3.0×, paving the way to energy-harvested printed systems. |
| Druh dokumentu: | Article |
| ISSN: | 1937-4151 0278-0070 |
| DOI: | 10.1109/tcad.2024.3524357 |
| Rights: | IEEE Copyright |
| Prístupové číslo: | edsair.doi.dedup.....8e231b0a3f17d3fdbd2fa0c9bd608d52 |
| Databáza: | OpenAIRE |
Full Text Finder 
Nájsť tento článok vo Web of Science | Abstrakt: | Printed electronics (PE) is an additive fabrication technology for manufacturing electronic circuits which not only allows for a highly flexible printing of arbitrary circuit patterns, but also produce soft, non-toxic, and degradable electronics at an extremely low cost. These properties are unmatched by siliconbased electronics, making PE an enabler of new application domains, e.g., fast moving consumer goods, wearables, and disposable healthcare devices. A particularly promising class of circuits in this technology is the printed analog neuromorphic circuits, offering efficient and highly tailored computational functionalities. In this work, we leverage the highly flexible fabrication process of PE to address the bottleneck of PE, i.e., the large feature sizes and low device counts. This issue is crucial, as it impairs the integration of printed circuits into target applications with limited footprint, such as smart bandaids. We also propose an evolutionary algorithm (EA) to improve the circuit compactness through circuit architecture optimization. As baseline, we compare the proposed EA method with a stateof-the-art pruning method and a modified area-aware pruning method. All of them are able to optimize circuit architecture alongside the component values of printed neuromorphic circuits. Experimental simulation reveals that the proposed EA approach can effectively achieve compact circuits and outperform the pruning method by 3.1× lower area with no loss of accuracy. As a byproduct, the power is reduced by 3.0×, paving the way to energy-harvested printed systems. |
|---|---|
| ISSN: | 19374151 02780070 |
| DOI: | 10.1109/tcad.2024.3524357 |