arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2312.02771cs.ETphysics.app-ph
收藏

利用磁畴壁物理扩展忆阻器蒙特卡洛

Scaling-up Memristor Monte Carlo with magnetic domain-wall physics

Thomas Dalgaty, Shogo Yamada, Anca Molnos, Eiji Kawasaki, Thomas Mesquida, François Rummens, Tatsuo Shibata, Yukihiro Urakawa, Yukio Terasaki, Tomoyuki Sasaki, Marc Duranton

AI总结:

本文将 SGLD 映射到磁畴壁忆阻器物理,提出推挽脉冲编程方法,把 MMC 规模提升五个数量级,并用 CIFAR-10 训练 ResNet18 验证近浮点性能。

摘要:

通过利用纳米级器件的内在随机性,Memristor Monte Carlo(MMC,忆阻器蒙特卡洛)是边缘学习系统的一种颇具前景的赋能技术。然而,由于多种算法和器件层面的限制,现有演示仅限于非常小的神经网络模型和数据集。我们讨论了这些限制,并阐述了如何通过将随机梯度朗之万动力学(SGLD)算法映射到磁畴壁忆阻器的物理特性上来克服它们,从而将 MMC 模型扩展五个数量级。我们提出了推挽脉冲编程方法,在物理层面实现 SGLD,并使用它在 CIFAR-10 数据集上训练基于畴壁的 ResNet18。在该任务上,我们观察到,在更新精度低至 6 至 7 位之间时,相对于浮点模型没有性能下降,表明我们已朝着利用有噪声模拟器件的大规模边缘学习系统迈出了一步。

英文摘要:

By exploiting the intrinsic random nature of nanoscale devices, Memristor Monte Carlo (MMC) is a promising enabler of edge learning systems. However, due to multiple algorithmic and device-level limitations, existing demonstrations have been restricted to very small neural network models and datasets. We discuss these limitations, and describe how they can be overcome, by mapping the stochastic gradient Langevin dynamics (SGLD) algorithm onto the physics of magnetic domain-wall Memristors to scale-up MMC models by five orders of magnitude. We propose the push-pull pulse programming method that realises SGLD in-physics, and use it to train a domain-wall based ResNet18 on the CIFAR-10 dataset. On this task, we observe no performance degradation relative to a floating point model down to an update precision of between 6 and 7-bits, indicating we have made a step towards a large-scale edge learning system leveraging noisy analogue devices.

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。

↑