Swift-CNN: Leveraging PCM Memory’s Fast Write Mode to Accelerate CNNs-Reference-Cited by-同舟云学术

Swift-CNN: Leveraging PCM Memory’s Fast Write Mode to Accelerate CNNs

Published:2023-12 Issue:4 Volume:15 Page:234-237
ISSN:1943-0663
Container-title:IEEE Embedded Systems Letters
language:
Short-container-title:IEEE Embedded Syst. Lett.

Author:

Siddhu Lokesh¹^ORCID,Nassar Hassan¹^ORCID,Bauer Lars¹^ORCID,Hakert Christian²^ORCID,Hölscher Nils²^ORCID,Chen Jian-Jia²^ORCID,Henkel Joerg¹^ORCID

Affiliation:

1. Chair for Embedded Systems, Karlsruhe Institute of Technology, Karlsruhe, Germany

2. Faculty of Informatics, Technical University of Dortmund, Dortmund, Germany

Funder

German Research Foundation (DFG): Reconfigurable Architectures and Real-Time Systems Co-Design for Non-Volatile Memory (part of “SPP 2377”) and NVM-OMA

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

Subject

General Computer Science,Control and Systems Engineering

Link

http://xplorestaging.ieee.org/ielx7/4563995/10330165/10194314.pdf?arnumber=10194314

Reference9 articles.

1. Group knowledge transfer: Federated learning of large CNNs at the edge;he;Proc NeurIPS,2020

2. Edge Machine Learning for AI-Enabled IoT Devices: A Review

3. Margin-Maximization in Binarized Neural Networks for Optimizing Bit Error Tolerance

4. Quick-and-Dirty: An Architecture for High-Performance Temporary Short Writes in MLC PCM

5. Exploiting Multiple Write Modes of Nonvolatile Main Memory in Embedded Systems

Cited by 1 articles. 订阅此论文施引文献订阅此论文施引文献，注册后可以免费订阅5篇论文的施引文献，订阅后可以查看论文全部施引文献

1. EDeN: Enabling Low-Power CNN Inference on Edge Devices Using Prefetcher-assisted NVM Systems;Proceedings of the 29th ACM/IEEE International Symposium on Low Power Electronics and Design;2024-08-05