An Efficient Binary Convolutional Neural Network with Numerous Skip Connections for Fog Computing

Fog computing is promising to solve the challenge caused by extremely large amount of data on cloud computing. In this study, an efficient binary convolutional neural network with numerous skip connections (BNSC-Net) is proposed for fog computing to enable real-time smart industrial applications. This network features decomposition convolution kernels and concatenated feature maps. Moreover, the network performance is further improved through expanding the update interval of Straight-Through-Estimator. To verify the performance, BNSC-Net is tested on two broadly used public datasets: ImageNet and CIFAR-10. Ablation study is firstly conducted to verify the effectiveness of the proposed improved operations, and results demonstrate that BNSC-Net can obviously increase the classification accuracy for both datasets. ImageNet based classification results indicate that BNSC-Net can achieve 59.9% TOP-1 accuracy that is 2.6% higher than the state-of-the-art binary neural networks, such as projection convolutional neural networks (PCNNs). Lastly, a subset with ten classes are selected from ImageNet to simulate the data collected in smart industry with limited categories, based on which BNSC-Net also demonstrate an impressive classification performance with friendly memory and calculation requirements. Particularly, the receiver operating characteristic curves of BNSC-Net surpass that of the state-of-the-art algorithm DeepIns. Therefore, the proposed BNSC-Net is effective and efficient for building deep learning-enabled industrial applications on fog nodes.