EncodingNet:一种基于编码的新型MAC设计,用于高效神经网络加速
EncodingNet: A Novel Encoding-based MAC Design for Efficient Neural Network Acceleration
AI总结:
提出基于编码的新型数字MAC设计,用逻辑门替代乘法器,缩短关键路径,减少流水线和寄存器,在保持精度的同时显著降低面积和功耗,经多个模型验证。
摘要:
深度神经网络(DNN)在图像分类和自然语言处理等许多领域取得了重大突破。然而,DNN的执行需要在硬件上进行大量的乘累加(MAC)操作,因此会产生较大的功耗。为了解决这一挑战,我们提出了一种基于编码的新型数字MAC设计。在这种新设计中,乘法器被简单的逻辑门所替代,以宽位表示形式来表示结果。新乘法器的输出通过按位加权累加进行相加,并且累加结果与现有的加速神经网络的计算平台兼容。由于乘法功能被简单的逻辑表示所取代,所得电路中的关键路径变得短得多。相应地,可以减少MAC阵列中用于存储部分和的流水线级和中间寄存器,从而显著减小面积并提高能效。所提出的设计已通过ResNet18-Cifar10、ResNet20-Cifar100、ResNet50-ImageNet、MobileNetV2-Cifar10、MobileNetV2-Cifar100和EfficientNetB0-ImageNet进行了综合和验证。实验结果证实,电路面积最多减少48.79%,执行DNN的功耗最多减少64.41%,同时神经网络的精度仍能很好地保持。本工作的开源代码可在GitHub上找到,链接为https://github.com/Bo-Liu-TUM/EncodingNet/。
英文摘要:
Deep neural networks (DNNs) have achieved great breakthroughs in many fields such as image classification and natural language processing. However, the execution of DNNs needs to conduct massive numbers of multiply-accumulate (MAC) operations on hardware and thus incurs a large power consumption. To address this challenge, we propose a novel digital MAC design based on encoding. In this new design, the multipliers are replaced by simple logic gates to represent the results with a wide bit representation. The outputs of the new multipliers are added by bit-wise weighted accumulation and the accumulation results are compatible with existing computing platforms accelerating neural networks. Since the multiplication function is replaced by a simple logic representation, the critical paths in the resulting circuits become much shorter. Correspondingly, pipelining stages and intermediate registers used to store partial sums in the MAC array can be reduced, leading to a significantly smaller area as well as better power efficiency. The proposed design has been synthesized and verified by ResNet18- Cifar10, ResNet20-Cifar100, ResNet50-ImageNet, MobileNetV2-Cifar10, MobileNetV2-Cifar100, and EfficientNetB0-ImageNet. The experimental results confirmed the reduction of circuit area by up to 48.79% and the reduction of power consumption of executing DNNs by up to 64.41%, while the accuracy of the neural networks can still be well maintained. The open source code of this work can be found on GitHub with link https://github.com/Bo-Liu-TUM/EncodingNet/.
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。