Knowledge Commons of Institute of Automation,CAS
Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimization | |
Weihan Chen![]() ![]() ![]() | |
2021-10 | |
会议名称 | International Conference on Computer Vision (ICCV) |
会议日期 | 2021-10-11 |
会议地点 | 线上举办 |
摘要 | Quantization is a widely used technique to compress and accelerate deep neural networks. However, conventional quantization methods use the same bit-width for all (or most of) the layers, which often suffer significant accuracy degradation in the ultra-low precision regime and ignore the fact that emergent hardware accelerators begin to support mixed-precision computation. Consequently, we present a novel and principled framework to solve the mixed-precision quantization problem in this paper. Briefly speaking, we first formulate the mixed-precision quantization as a discrete constrained optimization problem. Then, to make the optimization tractable, we approximate the objective function with second-order Taylor expansion and propose an efficient approach to compute its Hessian matrix. Finally, based on the above simplification, we show that the original problem can be reformulated as a Multiple-Choice Knapsack Problem (MCKP) and propose a greedy search algorithm to solve it efficiently. Compared with existing mixed-precision quantization works, our method is derived in a principled way and much more computationally efficient. Moreover, extensive experiments conducted on the |
收录类别 | EI |
七大方向——子方向分类 | AI芯片与智能计算 |
国重实验室规划方向分类 | 智能计算与学习 |
是否有论文关联数据集需要存交 | 否 |
文献类型 | 会议论文 |
条目标识符 | http://ir.ia.ac.cn/handle/173211/52065 |
专题 | 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 | Jian Cheng |
作者单位 | 1.NLPR & AIRIA, Institute of Automation, Chinese Academy of Sciences 2.School of Artificial Intelligence, University of Chinese Academy of Sciences |
推荐引用方式 GB/T 7714 | Weihan Chen,Peisong Wang,Jian Cheng. Towards Mixed-Precision Quantization of Neural Networks via Constrained Optimization[C],2021. |
条目包含的文件 | 下载所有文件 | |||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Chen_Towards_Mixed-P(696KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 下载 |
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