The Effect of Initial Weight, Learning Rate and Regularization on Generalization Performance and Efficiency

机译：初始权重，学习率和正则化对泛化性能和效率的影响

获取原文

获取原文并翻译 | 示例

页面导航

摘要
著录项
相似文献
相关主题

摘要

The goal of this paper is to study the factors that affect the generalization performance and efficiency for neural network learning. First, this paper will investigate the effect of initial weight ranges, learning rate, and regularization coefficient on generalization performance and learning speed. Based on this, we will propose a hybrid method that simultaneously considers these three factors, and dynamically tune the learning rate and regularization coefficient. Then we will present the results of some experimental comparison among these kinds of methods in several different problems. Finally, we will draw conclusions and make plan for future work.

机译：本文的目的是研究影响神经网络学习的泛化性能和效率的因素。首先，本文将研究初始权重范围，学习率和正则化系数对泛化性能和学习速度的影响。基于此，我们将提出一种混合方法，该方法同时考虑这三个因素，并动态调整学习率和正则化系数。然后，我们将介绍在几种不同问题中这些方法之间的一些实验比较结果。最后，我们将得出结论并制定未来工作的计划。

著录项

来源
《2002 6th International Conference on Signal Processing Proceedings (ICSP'02) Vol.2; Aug 26-30, 2002; Beijing, China》|2002年|p.1191-1194|共4页
会议地点 Beijing(CN);Beijing(CN)
作者
Wu Yan; Zhang Limimg;
展开▼
作者单位

Department of Computer Science and Engineering, Tongji University, Shanghai 200331;

展开▼
会议组织
原文格式 PDF
正文语种 eng
中图分类无线电电子学、电信技术;
关键词

相似文献

外文文献
中文文献
专利

1. Improving Generalization Performance of Natural Gradient Learning Using Optimized Regularization by NIC [J] . Hyeyoung Park, Noboru Murata, Shun-ichi Amari Neural computation . 2004,第2期

机译：使用NIC优化正则化来提高自然梯度学习的泛化性能
2. Impact of Different Random Initializations on Generalization Performance of Extreme Learning Machine [J] . Xiaofang Zhang, Xiaoli Lin, Rana Aamir Raza Ashfaq Journal of Computers . 2018,第7期

机译：不同随机初始化对极端学习机泛化性能的影响
3. Initializing receptive field-like weights in BP learning network (I)— Effects on convergence rate [J] . WANG Yunjiu, GUI He, QI Xianglin Progress in Natural Science . 1996,第2期

机译：在BP学习网络中初始化类似接收场的权重（I）-对收敛速度的影响
4. The effect of initial weight, learning rate and regularization on generalization performance and efficiency [C] . Wu Yan, Zhang Limimg . 2002

机译：初始权重，学习率和正则化对泛化性能和效率的影响
5. THE EFFECTS OF ISOMETRIC AND ISOTONIC ENDURANCE EXERCISES ON THE DEVELOPMENT OF CARDIOVASCULAR EFFICIENCY OF EIGHTH GRADE GIRLS CLASSIFIEDACCORDING TO INITIAL CARDIOVASCULAR EFFICIENCY AND WEIGHT. [D] . HAMILTON, XANDRA L. 1966

机译：等静力和等强度耐力对根据初始心血管效率和体重分类的八级女孩的心血管效率发展的影响。
6. A Robust and Accurate Deep-learning-based Method for the Segmentation of Subcortical Brain: Cross-dataset Evaluation of Generalization Performance [O] . Naoya Furuhashi, Shiho Okuhata, Tetsuo Kobayashi 2021

机译：基于深度学习的基于深度学习的基于深度学习的方法：泛化性能的交叉数据集评估
7. Does generalization performance of $l^q$ regularization learning depend on $q$? A negative example [O] . Lin, Shaobo, Xu, Chen, Zeng, Jingshan, 2013

机译：$ l ^ q $正则化学习的泛化表现是否依赖在$ q $？一个反面的例子
8. Neural Network Constructive Algorithms: Trading Generalization for Learning Efficiency. [R] . Smieja, F. J. 1992

机译：神经网络构造算法：学习效率的交易推广。

The Effect of Initial Weight, Learning Rate and Regularization on Generalization Performance and Efficiency

摘要

著录项

相似文献

相关主题

期刊订阅