Logo image
ANP-I: A 28-nm 1.5-pJ/SOP Asynchronous Spiking Neural Network Processor Enabling Sub-0.1-<inline-formula> <tex-math notation="LaTeX">$mu $</tex-math> </inline-formula>J/Sample On-Chip Learning for Edge-AI Applications
期刊文章   開放取用(OA)   同儕審查

ANP-I: A 28-nm 1.5-pJ/SOP Asynchronous Spiking Neural Network Processor Enabling Sub-0.1- $mu $ J/Sample On-Chip Learning for Edge-AI Applications

Jilin Zhang, Dexuan Huo, Jian Zhang, Chunqi Qian, Qi Liu, Liyang Pan, Zhihua Wang, Ning Qiao, Kea-Tiong TangHong Chen
IEEE Journal of Solid-State Circuits
2024

摘要

Application-specified integrated circuit (ASIC) asynchronous circuits Logic gates neuromorphic computing Neuromorphics Neurons on-chip learning Power demand spiking neural network (SNN) Synapses System-on-chip Timing Electrical and Electronic Engineering
Reducing learning energy consumption is critical to edge-artificial intelligence (AI) processors with on-chip learning since on-chip learning energy dominates energy consumption, especially for applications that require long-term learning. To achieve this goal, we optimize a neuromorphic learning algorithm and propose random target window (TW) selection, hierarchical update skip (HUS), and asynchronous time step acceleration (ATSA) to reduce the on-chip learning power consumption. Our approach results in a 28-nm 1.25-mm<inline-formula> <tex-math notation="LaTeX">$^{2}$</tex-math> </inline-formula> asynchronous neuromorphic processor (ANP-I) with on-chip learning energy per sample less than 15% of inference energy per sample. With all weights randomly initialized, this processor enables on-chip learning for edge-AI tasks such as gesture recognition, keyword spotting, and image classification, consuming sub-0.1 <inline-formula> <tex-math notation="LaTeX">$mu $</tex-math> </inline-formula>J of learning energy per sample at 0.56 V and 40-MHz frequency while maintaining <inline-formula> <tex-math notation="LaTeX">$>$</tex-math> </inline-formula>92% accuracy for all tasks.

檔案與連結 (1)

url
https://doi.org/10.1109/JSSC.2024.3357045檢視
已出版(紀錄版本) 開放

相關連結

指標

1 檢視次數

詳細資料

Logo image