A new stochastic computing multiplier with application to deep convolutional neural networks
A new stochastic computing multiplier with application to deep convolutional neural networks
复制标题
一种应用于深度卷积神经网络的新型随机计算乘法器
DOI：
--
复制
发表时间：
2017
期刊：
影响因子：
--
通讯作者：
Jongeun Lee
中科院分区：
文献类型：
--
作者：
H. Sim;Jongeun Lee
随机计算（SC）允许以极低的成本和低功耗实现常见算术运算。然而，固有的随机波动误差和长延迟的SC导致的准确性和能量效率的下降时，应用于卷积神经网络（CNN）。在本文中，我们解决了基于SC的CNN的两个关键问题，提出了一种新的SC乘法算法及其向量扩展，SC-MVM（矩阵向量乘法器），在这种情况下，一个SC乘法只需要几个周期，生成更准确的结果，并且可以以更低的成本实现，与传统的SC方法相比。我们使用为MNIST和CIFAR-10数据集设计的CNN进行的实验结果表明，我们基于SC的CNN不仅比传统的基于SC的CNN更准确，而且在计算方面比传统的基于SC的CNN高出40× 10490 × 10490倍，而且与相同精度的位宽优化定点实现相比，我们的CNN还可以实现更低的面积延迟积和更低的能量。
Stochastic computing (SC) allows for extremely low cost and low power implementations of common arithmetic operations. However inherent random fluctuation error and long latency of SC lead to the degradation of accuracy and energy efficiency when applied to convolutional neural networks (CNNs). In this paper we address the two critical problems of SC-based CNNs, by proposing a novel SC multiply algorithm and its vector extension, SC-MVM (Matrix-Vector Multiplier), under which one SC multiply takes just a few cycles, generates much more accurate results, and can be realized with significantly less cost, as compared to the conventional SC method. Our experimental results using CNNs designed for MNIST and CIFAR-10 datasets demonstrate that not only is our SC-based CNN more accurate and 40×∼490× more energy-efficient in computation than the conventional SC-based ones, but ours can also achieve lower area-delay product and lower energy compared with bitwidth-optimized fixed-point implementations of the same accuracy.