一种由SOT-MRAM存内计算系统实现的耐噪声、节省资源的概率二进制神经网络
A noise-tolerant, resource-saving probabilistic binary neural network implemented by the SOT-MRAM compute-in-memory system
AI总结:
本文提出基于SOT-MRAM存内计算的概率二进制神经网络,利用热涨落实现可控随机权重,在MNIST上以7.01%权重变化取得97.78%准确率,并将位级计算量降低6.9倍。
摘要:
我们报道了一种基于自旋轨道矩(SOT)磁阻随机存取存储器(MRAM)的概率二进制神经网络(PBNN),用于节省资源和硬件耐噪声的计算应用。在热涨落存在的情况下,非破坏性的SOT驱动磁化翻转特性产生了一个概率分布可控的随机权重矩阵。同时,所提出的存内计算(CIM)架构允许概率向量-矩阵乘法(PVMM)与二值化并发执行。此外,利用随机二进制单元传播多比特概率信息的有效性,我们基于SOT-MRAM的PBNN系统在MNIST数据库上,在7.01%的权重变化下,通过10个采样周期实现了97.78%的分类准确率,并且与全精度LeNet-5网络相比,位级计算操作数量减少了6.9倍。我们的工作为设计面向低功耗和有限计算资源应用的可靠神经网络提供了一个有说服力的框架。
英文摘要:
We report a spin-orbit torque(SOT) magnetoresistive random-access memory(MRAM)-based probabilistic binary neural network(PBNN) for resource-saving and hardware noise-tolerant computing applications. With the presence of thermal fluctuation, the non-destructive SOT-driven magnetization switching characteristics lead to a random weight matrix with controllable probability distribution. In the meanwhile, the proposed CIM architecture allows for the concurrent execution of the probabilistic vector-matrix multiplication (PVMM) and binarization. Furthermore, leveraging the effectiveness of random binary cells to propagate multi-bit probabilistic information, our SOT-MRAM-based PBNN system achieves a 97.78\% classification accuracy under a 7.01\% weight variation on the MNIST database through 10 sampling cycles, and the number of bit-level computation operations is reduced by a factor of 6.9 compared to that of the full-precision LeNet-5 network. Our work provides a compelling framework for the design of reliable neural networks tailored to the applications with low power consumption and limited computational resources.
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。