论文标题
基于浮点精度的有效神经网络训练的新型基于MRAM的过程中的内存加速器
A New MRAM-based Process In-Memory Accelerator for Efficient Neural Network Training with Floating Point Precision
论文作者
论文摘要
现代深度神经网络（DNN）的出色表现经常以艰巨的培训成本，限制了DNN创新的快速发展并引起了各种环境问题。为了降低培训的主要数据运动成本，过程中内存（PIM）已成为有前途的解决方案，因为它减轻了访问DNN权重的需求。但是，最新的PIM DNN训练加速器采用模拟/混合信号计算，该计算基于内存技术有限或数字计算，该内存技术支持有限的逻辑功能，因此需要复杂的程序才能实现浮点计算。在本文中，我们提出了一个基于旋转轨道扭矩磁随机访问记忆（SOT-MRAM）的数字PIM加速器，该数字PIM加速器支持浮点精度。具体而言，该新的加速器具有创新的（1）SOT-MRAM单元格，（2）完整的添加设计和（3）浮点计算。实验结果表明，与基于最新的基于PIM的DNN训练Accelerator相比，提出的基于SOT-MRAM PIM的DNN训练加速器分别可以实现3.3 $ \ times $，1.8 $ \ times $和2.5 $ \ times $ $改善，分别在能源，延迟和区域方面提高。
The excellent performance of modern deep neural networks (DNNs) comes at an often prohibitive training cost, limiting the rapid development of DNN innovations and raising various environmental concerns. To reduce the dominant data movement cost of training, process in-memory (PIM) has emerged as a promising solution as it alleviates the need to access DNN weights. However, state-of-the-art PIM DNN training accelerators employ either analog/mixed signal computing which has limited precision or digital computing based on a memory technology that supports limited logic functions and thus requires complicated procedure to realize floating point computation. In this paper, we propose a spin orbit torque magnetic random access memory (SOT-MRAM) based digital PIM accelerator that supports floating point precision. Specifically, this new accelerator features an innovative (1) SOT-MRAM cell, (2) full addition design, and (3) floating point computation. Experiment results show that the proposed SOT-MRAM PIM based DNN training accelerator can achieve 3.3$\times$, 1.8$\times$, and 2.5$\times$ improvement in terms of energy, latency, and area, respectively, compared with a state-of-the-art PIM based DNN training accelerator.
