ResNet-like Architecture with Low Hardware Requirements
This work addresses efficiency challenges for mobile and embedded devices in edge computing, but it is incremental as it adapts an existing neuron model to a more complex architecture.
The paper tackles the problem of high computational cost in deep neural network inference for edge devices by introducing a bipolar morphological ResNet (BM-ResNet) model, which reduces logic gate requirements by 2.1-2.9 times and latency by 15-30% with only a minor accuracy drop (e.g., from 99.3% to 99.1% on MNIST).
One of the most computationally intensive parts in modern recognition systems is an inference of deep neural networks that are used for image classification, segmentation, enhancement, and recognition. The growing popularity of edge computing makes us look for ways to reduce its time for mobile and embedded devices. One way to decrease the neural network inference time is to modify a neuron model to make it moreefficient for computations on a specific device. The example ofsuch a model is a bipolar morphological neuron model. The bipolar morphological neuron is based on the idea of replacing multiplication with addition and maximum operations. This model has been demonstrated for simple image classification with LeNet-like architectures [1]. In the paper, we introduce a bipolar morphological ResNet (BM-ResNet) model obtained from a much more complex ResNet architecture by converting its layers to bipolar morphological ones. We apply BM-ResNet to image classification on MNIST and CIFAR-10 datasets with only a moderate accuracy decrease from 99.3% to 99.1% and from 85.3% to 85.1%. We also estimate the computational complexity of the resulting model. We show that for the majority of ResNet layers, the considered model requires 2.1-2.9 times fewer logic gates for implementation and 15-30% lower latency.