Knowledge Commons of Institute of Automation,CAS
Rethinking the pid optimizer for stochastic optimization of deep networks | |
Shi, Lei1,2![]() ![]() ![]() ![]() | |
2020 | |
会议名称 | 2020 IEEE International Conference on Multimedia and Expo, ICME 2020 |
会议日期 | July 6, 2020 - July 10, 2020 |
会议地点 | London, United kingdom |
出版者 | IEEE Computer Society |
摘要 | Stochastic gradient descent with momentum (SGD-Momentum) always causes the overshoot problem due to the integral action of the momentum term. Recently, an ID optimizer is proposed to solve the overshoot problem with the help of derivative information. However, the derivative term suffers from the interference of the high-frequency noise, especially for the stochastic gradient descent method that uses minibatch data in each update step. In this work, we propose a complete PID optimizer, which weakens the effect of the D term and adds a P term to more stably alleviate the overshoot problem. To further reduce the interference of the high-frequency noise, two effective and efficient methods are proposed to stabilize the training process. Extensive experiments on three widely used benchmark datasets with different scales, i.e., MNIST, Cifar10 and TinyImageNet, demonstrate the superiority of our proposed PID optimizer on various popular deep neural networks. |
语种 | 英语 |
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 | 其他 |
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/173211/42212 |
专题 | 复杂系统认知与决策实验室_高效智能计算与学习 紫东太初大模型研究中心_图像与视频分析 |
通讯作者 | Shi, Lei |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences, NLPR AIRIA, China 2.School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing; 100049, China 3.State Grid Intelligence Technology Co. Ltd., China 4.CAS Center for Excellence in Brain Science and Intelligence Technology, China |
第一作者单位 | 模式识别国家重点实验室 |
通讯作者单位 | 模式识别国家重点实验室 |
推荐引用方式 GB/T 7714 | Shi, Lei,Zhang, Yifan,Wang, Wanguo,et al. Rethinking the pid optimizer for stochastic optimization of deep networks[C]:IEEE Computer Society,2020. |
条目包含的文件 | ||||||
文件名称/大小 | 文献类型 | 版本类型 | 开放类型 | 使用许可 | ||
Rethinking_the_PID_O(325KB) | 会议论文 | 开放获取 | CC BY-NC-SA | 浏览 |
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