標題: NV-BNN: An Accurate Deep Convolutional Neural Network Based on Binary STT-MRAM for Adaptive AI Edge
作者: Chang, Chih-Chcng
Wu, Ming-Hung
Lin, Jia-Wei
Li, Chun-Hsien
Parmar, Vivek
Lee, Heng-Yuan
Wei, Jeng-Hua
Sheu, Shyh-Shyuan
Suri, Manan
Chang, Tian-Sheuan
Hou, Tuo-Hung
電子工程學系及電子研究所
Department of Electronics Engineering and Institute of Electronics
公開日期: 1-Jan-2019
摘要: Binary STT-MRAM is a highly anticipated embedded non-volatile memory technology in advanced logic nodes <28 nm. How to enable its in-memory computing (IMC) capability is critical for enhancing AI Edge. Based on the soon-available STT-MRAM, we report the first binary deep convolutional neural network (NV-BNN) capable of both local and remote learning. Exploiting intrinsic cumulative switching probability, accurate online training of CIFAR-10 color images (similar to 90%) is realized using a relaxed endurance spec (switching <= 20 times) and hybrid digital/IMC design. For offline training, the accuracy loss due to imprecise weight placement can be mitigated using a rapid non-iterative training-with-noise and fine-tuning scheme.
URI: http://dx.doi.org/10.1145/3316781.3317872
http://hdl.handle.net/11536/152982
ISBN: 978-1-4503-6725-7
DOI: 10.1145/3316781.3317872
期刊: PROCEEDINGS OF THE 2019 56TH ACM/EDAC/IEEE DESIGN AUTOMATION CONFERENCE (DAC)
起始頁: 0
結束頁: 0
Appears in Collections:Conferences Paper