HarDNet: A Low Memory Traffic Network
arXiv:1909.00948
Abstract
State-of-the-art neural network architectures such as ResNet, MobileNet, and DenseNet have achieved outstanding accuracy over low MACs and small model size counterparts. However, these metrics might not be accurate for predicting the inference time. We suggest that memory traffic for accessing intermediate feature maps can be a factor dominating the inference latency, especially in such tasks as real-time object detection and semantic segmentation of high-resolution video. We propose a Harmonic Densely Connected Network to achieve high efficiency in terms of both low MACs and memory traffic. The new network achieves 35%, 36%, 30%, 32%, and 45% inference time reduction compared with FC-DenseNet-103, DenseNet-264, ResNet-50, ResNet-152, and SSD-VGG, respectively. We use tools including Nvidia profiler and ARM Scale-Sim to measure the memory traffic and verify that the inference latency is indeed proportional to the memory traffic consumption and the proposed network consumes low memory traffic. We conclude that one should take memory traffic into consideration when designing neural network architectures for high-resolution applications at the edge.
arXiv commentICCV 2019
References in corpus (10)
- SqueezeNet: AlexNet-level accuracy with 50x fewer parameters and <0.5MB model size
- Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
- Training Very Deep Networks
- Convolutional Neural Networks using Logarithmic Data Representation
- NullHop: A Flexible Convolutional Neural Network Accelerator Based on Sparse Representations of Feature Maps
- XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks
- Hardware-oriented Approximation of Convolutional Neural Networks
- SCALE-Sim: Systolic CNN Accelerator Simulator
- The Importance of Skip Connections in Biomedical Image Segmentation
- Log-DenseNet: How to Sparsify a DenseNet