| Samples Selection for Artificial Neural Network Training in Preliminary Structural Design |
| |
| 作者姓名： | 童霏 刘西拉 |
| |
| 作者单位： | [1]DepartmentofCivilEngineering,TsinghuaUniversity,Beijing100084,China [2]DepartmentofCivilEngineering,ShanghaiJiaoTongUniversity,Shanghai200030,China |
| |
| 摘 要： | An artificial neural network (ANN) is applied in the preliminary structural design of reticulated shells. Major efforts are made to enhance the generalization ability of networks through well-selected training samples. Number-theoretic methods (NTMs) are adopted to generate samples with low discrepancy, i.e.uniformly scattered in the domain, where discrepancy is a quantitative measurement of the uniformity. The discrepancy of the NTM-based sample set is 1/6-1/7 that of samples with equal spacing. In a case study,networks trained by NTM-based samples are compared with those trained by equal-spaced samples in generalizing performance. The results show that both the computational precision and stability of the former ANNs are more satisfactory than those of the latter. It is concluded that the flexibility of ANNs in generalizing can be effectively increased by use of uniformly distributed training samples rather than simply piling data.More reliable uniformity should be obtained, however, through NTMs instead of equal-spaced samples.
|
| 关 键 词： | 人造神经网络系统 结构设计 建筑设计 数论 稳定性分析 |
Samples Selection for Artificial Neural Network Training in Preliminary Structural Design |
| |
| TONG Fei,LIU Xila.Samples Selection for Artificial Neural Network Training in Preliminary Structural Design[J].Tsinghua Science and Technology,2005,10(2):233-239. |
| |
| Authors: | TONG Fei LIU Xila |
| |
| Institution: | TONG Fei,LIU Xila Department of Civil Engineering,Tsinghua University,Beijing 100084,China, Department of Civil Engineering,Shanghai Jiao Tong University,Shanghai 200030,China |
| |
| Abstract: | An artificial neural network (ANN) is applied in the preliminary structural design of reticulated shells. Major efforts are made to enhance the generalization ability of networks through well-selected train- ing samples. Number-theoretic methods (NTMs) are adopted to generate samples with low discrepancy, i.e., uniformly scattered in the domain, where discrepancy is a quantitative measurement of the uniformity. The discrepancy of the NTM-based sample set is 1/6-1/7 that of samples with equal spacing. In a case study, networks trained by NTM-based samples are compared with those trained by equal-spaced samples in generalizing performance. The results show that both the computational precision and stability of the former ANNs are more satisfactory than those of the latter. It is concluded that the flexibility of ANNs in generalizing can be effectively increased by use of uniformly distributed training samples rather than simply piling data. More reliable uniformity should be obtained, however, through NTMs instead of equal-spaced samples. |
| |
| Keywords: | artificial neural networks number-theoretic methods preliminary structural design reticulated shell |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
|
