---
_id: '67472'
abstract:
- lang: eng
text: This paper explores the challenges and opportunities of integrating non-volatile
memories (NVMs) into embedded systems for machine learning. NVMs offer advantages
such as increased memory density, lower power consumption, non-volatility, and
compute-in- memory capabilities. The paper focuses on integrating NVMs into embedded
systems, particularly in intermittent computing, where systems operate during
periods of available energy. NVM technologies bring persistence closer to the
CPU core, enabling efficient designs for energy-constrained scenarios. Next, computation
in resistive NVMs is explored, highlighting its potential for accelerating machine
learning algorithms. However, challenges related to reliability and device non-idealities
need to be addressed. The paper also discusses memory-centric machine learning,
leveraging NVMs to overcome the memory wall challenge. By optimizing memory layouts
and utilizing probabilistic decision tree execution and neural network sparsity,
NVM-based systems can improve cache behavior and reduce unnecessary computations.
In conclusion, the paper emphasizes the need for further research and optimization
for the widespread adoption of NVMs in embedded systems presenting relevant challenges,
especially for machine learning applications.
author:
- first_name: Jörg
full_name: Henkel, Jörg
last_name: Henkel
- first_name: Lokesh
full_name: Siddhu, Lokesh
last_name: Siddhu
- first_name: Lars
full_name: Bauer, Lars
last_name: Bauer
- first_name: Jürgen
full_name: Teich, Jürgen
last_name: Teich
- first_name: Stefan
full_name: Wildermann, Stefan
last_name: Wildermann
- first_name: Mehdi
full_name: Tahoori, Mehdi
last_name: Tahoori
- first_name: Mahta
full_name: Mayahinia, Mahta
last_name: Mayahinia
- first_name: Jeronimo
full_name: Castrillon, Jeronimo
last_name: Castrillon
- first_name: Asif Ali
full_name: Khan, Asif Ali
last_name: Khan
- first_name: Hamid
full_name: Farzaneh, Hamid
last_name: Farzaneh
- first_name: João Paulo C.
full_name: de Lima, João Paulo C.
last_name: de Lima
- first_name: Jian-Jia
full_name: Chen, Jian-Jia
last_name: Chen
- first_name: Christian
full_name: Hakert, Christian
last_name: Hakert
- first_name: Kuan-Hsun
full_name: Chen, Kuan-Hsun
last_name: Chen
- first_name: Chia-Lin
full_name: Yang, Chia-Lin
last_name: Yang
- first_name: Hsiang-Yun
full_name: Cheng, Hsiang-Yun
last_name: Cheng
citation:
ama: 'Henkel J, Siddhu L, Bauer L, et al. Special Session – Non-Volatile Memories:
Challenges and Opportunities for Embedded System Architectures with Focus on Machine
Learning Applications. In: Proceedings of the 2023 International Conference
on Compilers, Architecture, and Synthesis of Embedded Systems (CASES). CASES
’23 Companion. ; 2023:11–20. doi:10.1145/3607889.3609088'
apa: 'Henkel, J., Siddhu, L., Bauer, L., Teich, J., Wildermann, S., Tahoori, M.,
Mayahinia, M., Castrillon, J., Khan, A. A., Farzaneh, H., de Lima, J. P. C., Chen,
J.-J., Hakert, C., Chen, K.-H., Yang, C.-L., & Cheng, H.-Y. (2023). Special
Session – Non-Volatile Memories: Challenges and Opportunities for Embedded System
Architectures with Focus on Machine Learning Applications. Proceedings of the
2023 International Conference on Compilers, Architecture, and Synthesis of Embedded
Systems (CASES), 11–20. https://doi.org/10.1145/3607889.3609088'
bibtex: '@inproceedings{Henkel_Siddhu_Bauer_Teich_Wildermann_Tahoori_Mayahinia_Castrillon_Khan_Farzaneh_et
al._2023, series={CASES ’23 Companion}, title={Special Session – Non-Volatile
Memories: Challenges and Opportunities for Embedded System Architectures with
Focus on Machine Learning Applications}, DOI={10.1145/3607889.3609088},
booktitle={Proceedings of the 2023 International Conference on Compilers, Architecture,
and Synthesis of Embedded Systems (CASES)}, author={Henkel, Jörg and Siddhu, Lokesh
and Bauer, Lars and Teich, Jürgen and Wildermann, Stefan and Tahoori, Mehdi and
Mayahinia, Mahta and Castrillon, Jeronimo and Khan, Asif Ali and Farzaneh, Hamid
and et al.}, year={2023}, pages={11–20}, collection={CASES ’23 Companion} }'
chicago: 'Henkel, Jörg, Lokesh Siddhu, Lars Bauer, Jürgen Teich, Stefan Wildermann,
Mehdi Tahoori, Mahta Mayahinia, et al. “Special Session – Non-Volatile Memories:
Challenges and Opportunities for Embedded System Architectures with Focus on Machine
Learning Applications.” In Proceedings of the 2023 International Conference
on Compilers, Architecture, and Synthesis of Embedded Systems (CASES), 11–20.
CASES ’23 Companion, 2023. https://doi.org/10.1145/3607889.3609088.'
ieee: 'J. Henkel et al., “Special Session – Non-Volatile Memories: Challenges
and Opportunities for Embedded System Architectures with Focus on Machine Learning
Applications,” in Proceedings of the 2023 International Conference on Compilers,
Architecture, and Synthesis of Embedded Systems (CASES), 2023, pp. 11–20,
doi: 10.1145/3607889.3609088.'
mla: 'Henkel, Jörg, et al. “Special Session – Non-Volatile Memories: Challenges
and Opportunities for Embedded System Architectures with Focus on Machine Learning
Applications.” Proceedings of the 2023 International Conference on Compilers,
Architecture, and Synthesis of Embedded Systems (CASES), 2023, pp. 11–20,
doi:10.1145/3607889.3609088.'
short: 'J. Henkel, L. Siddhu, L. Bauer, J. Teich, S. Wildermann, M. Tahoori, M.
Mayahinia, J. Castrillon, A.A. Khan, H. Farzaneh, J.P.C. de Lima, J.-J. Chen,
C. Hakert, K.-H. Chen, C.-L. Yang, H.-Y. Cheng, in: Proceedings of the 2023 International
Conference on Compilers, Architecture, and Synthesis of Embedded Systems (CASES),
2023, pp. 11–20.'
date_created: 2026-10-09T08:51:06Z
date_updated: 2026-10-21T09:14:11Z
doi: 10.1145/3607889.3609088
page: 11–20
publication: Proceedings of the 2023 International Conference on Compilers, Architecture,
and Synthesis of Embedded Systems (CASES)
publication_identifier:
isbn:
- '9798400702907'
issn:
- 2643-1726
series_title: CASES ’23 Companion
status: public
title: 'Special Session – Non-Volatile Memories: Challenges and Opportunities for
Embedded System Architectures with Focus on Machine Learning Applications'
type: conference
user_id: '131313'
year: '2023'
...