| ZHANG Qingsong,SUN Linjun,YANG Guowei,LU Baoli,NING Xin,LI Weijun.TBNN: totally-binary neural network for image classification[J].Optoelectronics Letters,2023,(2):117-122 |
| TBNN: totally-binary neural network for image classification |
| Author Name | Affiliation | | ZHANG Qingsong | School of Electronic Information, Qingdao University, Qingdao 266071, China
Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China | | SUN Linjun | Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China | | YANG Guowei | School of Electronic Information, Qingdao University, Qingdao 266071, China | | LU Baoli | Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China | | NING Xin | Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China | | LI Weijun | Institute of Semiconductors, Chinese Academy of Sciences, Beijing 100083, China |
|
| Abstract: |
| Most binary networks apply full precision convolution at the first layer. Changing the first layer to the binary convolution will result in a significant loss of accuracy. In this paper, we propose a new approach to solve this problem by widening the data channel to reduce the information loss of the first convolutional input through the sign function. In addition, widening the channel increases the computation of the first convolution layer, and the problem is solved by using group convolution. The experimental results show that the accuracy of applying this paper's method to state-of-the-art (SOTA) binarization method is significantly improved, proving that this paper's method is effective and feasible. |
| Hits: 325 |
| Download times: 0 |
|
| View Full Text Download reader |
|
|
|
