Knowledge Commons of Institute of Automation,CAS
Hardware acceleration of CNN with one-hot quantization of weights and activations | |
Li, Gang1,2; Wang, Peisong1,2; Liu, Zejian1,2; Leng, Cong1,2; Cheng, Jian1,2 | |
2020 | |
会议名称 | Design, Automation & Test in Europe Conference & Exhibition (DATE) |
会议日期 | 2020-3 |
会议地点 | Grenoble, France |
摘要 | In this paper, we propose a novel one-hot representation for weights and activations in CNN model and demonstrate its benefits on hardware accelerator design. Specifically, rather than merely reducing the bitwidth, we quantize both weights and activations into n-bit integers that containing only one non-zero bit per value. In this way, the massive multiply and accumulates (MACs) are equivalent to additions of powers of two that can be efficiently calculated with histogram based computations. Experiments on the ImageNet classification task show that comparable accuracy can be obtained on our proposed One-Hot Networks (OHN) compared to conventional fixed-point networks. As case studies, we evaluate the efficacy of the one-hot data representation on two state-of-the-art CNN accelerators on FPGA, our preliminary results show that 50% and 68.5% resource saving can be achieved on DaDianNao and Laconic respectively. Besides, the one-hot optimized Laconic can further achieve an average speedup of 4.94× on AlexNet. |
收录类别 | EI |
七大方向——子方向分类 | 图像视频处理与分析 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/40617 |
专题 | 紫东太初大模型研究中心_图像与视频分析 复杂系统认知与决策实验室_高效智能计算与学习 |
通讯作者 | Cheng, Jian |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences 2.University of Chinese Academy of Sciences |
第一作者单位 | 中国科学院自动化研究所 |
通讯作者单位 | 中国科学院自动化研究所 |
推荐引用方式 GB/T 7714 | Li, Gang,Wang, Peisong,Liu, Zejian,et al. Hardware acceleration of CNN with one-hot quantization of weights and activations[C],2020. |
条目包含的文件 | 条目无相关文件。 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论