A Novel Design of Adaptive and Hierarchical Convolutional Neural Networks using Partial Reconfiguration on FPGA
A Novel Design of Adaptive and Hierarchical Convolutional Neural Networks using Partial Reconfiguration on FPGA
复制标题
DOI：
10.1109/hpec.2019.8916237
复制
发表时间：
2019-09
期刊：
影响因子：
--
通讯作者：
Mohammad Farhadi;Mehdi Ghasemi;Yezhou Yang
中科院分区：
文献类型：
--
作者：
Mohammad Farhadi;Mehdi Ghasemi;Yezhou Yang
目前，基于卷积神经网络(CNN)的视觉识别研究大多遵循“模型越深，置信度越高”的思想，以获得更高的识别精度。同时，模型越深，计算量越大。另一方面，对于大量的识别挑战，系统可以使用简单的模型或所谓的浅层网络正确地对图像进行分类。此外，CNN的实现还面临着对嵌入式设备的尺寸、重量和能量的限制。在本文中，我们在资源受限的MPSoC上用CPU和FPGA实现了浅层和深层网络之间的自适应切换，以达到最高的吞吐量。为此，我们开发并提出了一种新的CNN体系结构，其中GATE决定使用更深层次的模型是否有益。由于现场可编程门阵列的资源有限，采用了部分可重构的思想来适应在现场可编程门阵列资源上的深层神经网络。我们报告了在CIFAR-10、CIFAR-100和SVHN数据集上的实验结果，以验证我们的方法。使用置信度作为决策因素，CIFAR10、CIFAR-100和SVHN在最深层网络中的计算量分别只有69.8%、71.8%和43.8%，而对于SVHN数据集，它可以保持预期的精度，吞吐量约为400幅/秒。Https://github.com/mfarhadi/AHCNN.
Nowadays most research in visual recognition using Convolutional Neural Networks (CNNs) follows the “deeper model with deeper confidence” belief to gain a higher recognition accuracy. At the same time, deeper model brings heavier computation. On the other hand, for a large chunk of recognition challenges, a system can classify images correctly using simple models or so-called shallow networks. Moreover, the implementation of CNNs faces with the size, weight, and energy constraints on the embedded devices. In this paper, we implement the adaptive switching between shallow and deep networks to reach the highest throughput on a resource-constrained MPSoC with CPU and FPGA. To this end, we develop and present a novel architecture for the CNNs where a gate makes the decision whether using the deeper model is beneficial or not. Due to resource limitation on FPGA, the idea of partial reconfiguration has been used to accommodate deep CNNs on the FPGA resources. We report experimental results on CIFAR-10, CIFAR-100, and SVHN datasets to validate our approach. Using confidence metric as the decision making factor, only 69.8%, 71.8%, and 43.8% of the computation in the deepest network is done for CIFAR10, CIFAR-100, and SVHN while it can maintain the desired accuracy with the throughput of around 400 images per second for SVHN dataset. https://github.com/mfarhadi/AHCNN.