A Survey on Acceleration of Deep Convolutional Neural Networks
share
Deep Neural Networks have achieved remarkable progress during the past few years and are currently the fundamental tools of many intelligent systems. At the same time, the computational complexity and resource consumption of these networks are also continuously increasing. This will pose a significant challenge to the deployment of such networks, especially for real-time applications or on resource-limited devices. Thus, network acceleration have become a hot topic within the deep learning community. As for hardware implementation of deep neural networks, a batch of accelerators based on FPGA/ASIC have been proposed these years. In this paper, we provide a comprehensive survey about the recent advances on network acceleration, compression and accelerator design from both algorithm and hardware side. Specifically, we provide thorough analysis for each of the following topics: network pruning, low-rank approximation, network quantization, teacher-student networks, compact network design and hardware accelerator. Finally, we make a discussion and introduce a few possible future directions.
READ FULL TEXT