arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2412.03089cs.ETcs.AR
收藏

深度学习加速器中ReRAM交叉阵列的在线软错误容忍

Online Soft Error Tolerance in ReRAM Crossbars for Deep Learning Accelerators

Benyamin Khezeli, Hamid Reza Zarandi, Elham Cheshmikhani

AI总结:

针对ReRAM交叉阵列因非理想因素导致PIM加速器精度下降问题,提出基于测试输入向量和ECCs的在线软错误检测校正方法,在MNIST和CIFAR-10数据集上实现近无故障精度且开销更低。

摘要:

阻变随机存取存储器(ReRAM)交叉阵列是原位矩阵向量乘法(MVM)的理想候选,这是深度学习算法中的常见操作。尽管有优势,这些新兴非易失性存储器因制造工艺不成熟和运行时错误等非理想因素易出错,导致存内处理(PIM)加速器精度下降。本文提出ReRAM交叉阵列的在线软错误检测与校正方法,利用测试输入向量和纠错码(ECCs)检测并校正故障列。该方法在MNIST和CIFAR-10数据集上实现接近无故障的神经网络(NNs)精度,且与近期方法相比面积开销和功耗更低。

英文摘要:

Resistive Random-Access Memory (ReRAM) crossbar arrays are promising candidates for in-situ matrix-vector multiplication (MVM), a frequent operation in Deep Learning algorithms. Despite their advantages, these emerging non-volatile memories are susceptible to errors due to non-idealities such as immature fabrication processes and runtime errors, which lead to accuracy degradation in Processing-in-Memory (PIM) accelerators. This paper proposes an online soft error detection and correction method in ReRAM crossbar arrays. We utilize a test input vector and Error Correcting Codes (ECCs) to detect and correct faulty columns. The proposed approach demonstrates near fault-free accuracy for Neural Networks (NNs) on MNIST and CIFAR-10 datasets, with low area overhead and power consumption compared to recent methods.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑