当前位置: X-MOL 学术arXiv.cs.AR › 论文详情
Our official English website, www.x-mol.net, welcomes your feedback! (Note: you will need to create a separate account there.)
ReSpawn: Energy-Efficient Fault-Tolerance for Spiking Neural Networks considering Unreliable Memories
arXiv - CS - Hardware Architecture Pub Date : 2021-08-23 , DOI: arxiv-2108.10271
Rachmad Vidya Wicaksana Putra, Muhammad Abdullah Hanif, Muhammad Shafique

Spiking neural networks (SNNs) have shown a potential for having low energy with unsupervised learning capabilities due to their biologically-inspired computation. However, they may suffer from accuracy degradation if their processing is performed under the presence of hardware-induced faults in memories, which can come from manufacturing defects or voltage-induced approximation errors. Since recent works still focus on the fault-modeling and random fault injection in SNNs, the impact of memory faults in SNN hardware architectures on accuracy and the respective fault-mitigation techniques are not thoroughly explored. Toward this, we propose ReSpawn, a novel framework for mitigating the negative impacts of faults in both the off-chip and on-chip memories for resilient and energy-efficient SNNs. The key mechanisms of ReSpawn are: (1) analyzing the fault tolerance of SNNs; and (2) improving the SNN fault tolerance through (a) fault-aware mapping (FAM) in memories, and (b) fault-aware training-and-mapping (FATM). If the training dataset is not fully available, FAM is employed through efficient bit-shuffling techniques that place the significant bits on the non-faulty memory cells and the insignificant bits on the faulty ones, while minimizing the memory access energy. Meanwhile, if the training dataset is fully available, FATM is employed by considering the faulty memory cells in the data mapping and training processes. The experimental results show that, compared to the baseline SNN without fault-mitigation techniques, ReSpawn with a fault-aware mapping scheme improves the accuracy by up to 70% for a network with 900 neurons without retraining.

中文翻译:

ReSpawn:考虑不可靠内存的尖峰神经网络的节能容错

尖峰神经网络 (SNN) 由于其受生物启发的计算而显示出具有低能量和无监督学习能力的潜力。但是,如果在存储器中存在硬件引起的故障(可能来自制造缺陷或电压引起的近似误差)的情况下执行它们的处理,则它们的精度可能会下降。由于最近的工作仍然集中在 SNN 中的故障建模和随机故障注入,SNN 硬件架构中的内存故障对准确性的影响和各自的故障缓解技术没有得到彻底的探讨。为此,我们提出了 ReSpawn,这是一种新颖的框架,用于减轻片外和片上存储器中故障对弹性和节能 SNN 的负面影响。ReSpawn 的关键机制是:(1)分析SNNs的容错性;(2) 通过 (a) 内存中的故障感知映射 (FAM) 和 (b) 故障感知训练和映射 (FATM) 提高 SNN 容错能力。如果训练数据集不完全可用,FAM 将通过有效的位改组技术使用,该技术将有效位放在非故障存储单元上,将不重要位放在故障存储单元上,同时最大限度地减少内存访问能量。同时,如果训练数据集完全可用,则在数据映射和训练过程中考虑到错误的内存单元来采用 FATM。实验结果表明,与没有故障缓解技术的基线 SNN 相比,具有故障感知映射方案的 ReSpawn 将具有 900 个神经元的网络的准确度提高了 70%,而无需重新训练。(2) 通过 (a) 内存中的故障感知映射 (FAM) 和 (b) 故障感知训练和映射 (FATM) 提高 SNN 容错能力。如果训练数据集不完全可用,FAM 将通过有效的位改组技术使用,该技术将有效位放在非故障存储单元上,将不重要位放在故障存储单元上,同时最大限度地减少内存访问能量。同时,如果训练数据集完全可用,则在数据映射和训练过程中考虑到错误的内存单元来采用 FATM。实验结果表明,与没有故障缓解技术的基线 SNN 相比,具有故障感知映射方案的 ReSpawn 将具有 900 个神经元的网络的准确度提高了 70%,而无需重新训练。(2) 通过 (a) 内存中的故障感知映射 (FAM) 和 (b) 故障感知训练和映射 (FATM) 提高 SNN 容错能力。如果训练数据集不完全可用,FAM 将通过有效的位改组技术使用,该技术将有效位放在非故障存储单元上,将不重要位放在故障存储单元上,同时最大限度地减少内存访问能量。同时,如果训练数据集完全可用,则在数据映射和训练过程中考虑到错误的内存单元来采用 FATM。实验结果表明,与没有故障缓解技术的基线 SNN 相比,具有故障感知映射方案的 ReSpawn 将具有 900 个神经元的网络的准确度提高了 70%,而无需重新训练。
更新日期:2021-08-24
down
wechat
bug