Abstract | ||
---|---|---|
Convolutional neural networks(CNNs) have been widely applied in various applications. However, the computation-intensive convolutional layers and memory-intensive fully connected layers have brought many challenges to the implementation of CNN on embedded platforms. To overcome this problem, this work proposes a power-efficient accelerator for CNNs, and different methods are applied to optimize the convolutional layers and fully connected layers. For the convolutional layer, the accelerator first rearranges the input features into matrix on-the-fly when storing them to the on-chip buffers. Thus the computation of convolutional layer can be completed through matrix multiplication. For the fully connected layer, the batch-based method is used to reduce the required memory bandwidth, which also can be completed through matrix multiplication. Then a two-layer pipelined computation method for matrix multiplication is proposed to increase the throughput. As a case study, we implement a widely used CNN model, LeNet-5, on an embedded device. It can achieve a peak performance of 34.48 GOP/s and the power efficiency with the value of 19.45 GOP/s/W under 100MHz clock frequency which outperforms previous approaches. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/CLUSTER.2017.47 | 2017 IEEE International Conference on Cluster Computing (CLUSTER) |
Keywords | Field | DocType |
power-efficient accelerator,computation-intensive convolutional layers,convolutional layer,matrix multiplication,computation method,convolutional neural networks,CNN,memory intensive fully connected layers,on chip buffers,embedded device.,frequency 100.0 MHz | Memory bandwidth,System on a chip,Computer science,Convolutional neural network,Parallel computing,Memory management,Throughput,Artificial neural network,Matrix multiplication,Clock rate | Conference |
ISSN | ISBN | Citations |
1552-5244 | 978-1-5386-2327-5 | 2 |
PageRank | References | Authors |
0.38 | 5 | 8 |
Name | Order | Citations | PageRank |
---|---|---|---|
Fan Sun | 1 | 6 | 3.27 |
Chao Wang | 2 | 372 | 62.24 |
Lei Gong | 3 | 65 | 13.52 |
Chongchong Xu | 4 | 7 | 4.63 |
Yiwei Zhang | 5 | 52 | 12.65 |
Yuntao Lu | 6 | 6 | 2.59 |
Xi Li | 7 | 202 | 36.61 |
Xuehai Zhou | 8 | 551 | 77.54 |