STRIDe：跨耦合STT-MRAM实现深度神经网络加速器的鲁棒性内存计算
STRIDe: Cross-Coupled STT-MRAM Enabling Robust In-Memory-Computing for Deep Neural Network Accelerators
AI总结：
针对STT-MRAM在内存计算中因低区分度导致的鲁棒性问题，提出STRIDe方案，通过跨耦合提升比特单元电流比，提升XNOR-IMC和AND-IMC的感测边距和读扰边距，实现高精度推理。
摘要：
随着深度神经网络（DNN）模型的指数级增长，其在资源受限的边缘平台上的部署变得越来越具有挑战性。在内存计算（IMC）中使用非易失性存储器（NVMs）已成为一种潜在的解决方案，因其比标准DNN硬件平台具有更高的能效。在各种NVMs中，STT-MRAM因其高耐久性和其他优势而极具前景。然而，其IMC实现具有挑战性，因为其本质上具有低区分度。这个问题由于阵列非理想性和工艺变化而加剧，导致IMC鲁棒性差和严重的推理准确性下降。为了解决这个问题，我们提出了STRIDe - 基于STT-MRAM的IMC，利用交叉耦合作用来提升比特单元级别的高到低电流比，达到8000。我们提出了两种STRIDe设计变种，两者都为输入和权重在{-1, 1}(XNOR-IMC)和{0, 1}(AND-IMC)范围内提供鲁棒的IMC。我们的STRIDe阵列评估显示，XNOR-IMC和AND-IMC的感测边距（SM）分别提高了3.86倍和1.77倍，且比标准MRAM-IMC设计提高了27.6%的读扰边距（RDM）。STRIDe的增强鲁棒性在考虑交叉bar非理想性和工艺变化时，为ResNet18 BNN和在CIFAR10数据集上训练的4位DNN实现了接近软件的推理精度。我们观察到，与标准MRAM设计相比，BNN的精度提高了70%，4位DNN的精度提高了35%，尽管伴随着一些能量-面积-延迟的惩罚。
英文摘要：
As deep neural network (DNN) models are growing exponentially in size, their deployment on resource-constrained edge platforms is becoming increasingly challenging. In-memory-computing (IMC) with non-volatile memories (NVMs) has emerged as a potential solution by virtue of its higher energy efficiency compared to standard DNN hardware platforms. Amongst various NVMs, STT-MRAM is highly promising owing to its high endurance and other benefits. However, their IMC implementation is challenging because of their inherently low distinguishability. This issue is exacerbated due to array non-idealities and process-variations, leading to poor IMC robustness and severe inference accuracy degradation. To address this problem, we propose STRIDe - STT-MRAM-based IMC leveraging cross-coupling action to boost the bitcell-level high-to-low current ratio to up to 8000. We propose two flavors of STRIDe designs, both offering robust IMC for inputs and weights in {-1, 1}(XNOR-IMC) and {0, 1}(AND-IMC) regime. Our evaluations for STRIDe arrays show up to 3.86x and 1.77x sense margin (SM) improvement for XNOR-IMC and AND-IMC, respectively, and up to 27.6% read disturb margin (RDM) improvement over standard MRAM-IMC designs. The enhanced robustness of STRIDe translates to near-software inference accuracies (considering crossbar non-idealities and process variations) for ResNet18 BNN and 4-bit DNN trained on CIFAR10 dataset. We observe accuracy improvements of up to 70% (for BNN) and up to 35%(for 4-bit DNN) over standard MRAM designs, albeit with some energy-area-latency penalty.
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。
