| 基于神经网络测试码生成的一个鲁棒算法 |
| |
| 引用本文： | 陈朝阳,丁明跃.基于神经网络测试码生成的一个鲁棒算法[J].华中科技大学学报(自然科学版),1999(9). |
| |
| 作者姓名： | 陈朝阳 丁明跃 |
| |
| 作者单位： | 华中理工大学图像识别与人工智能研究所图像信息处理与智能控制国家教育部开放实验室 |
| |
| 摘 要： | 针对基于组合电路神经网络模型进行测试码生成，提出了一个对神经元连接权值不敏感的鲁棒算法——改进的遗传算法．它克服了运用概率松驰搜索算法求解测试矢量时，神经元之间连接权值的选择影响测试结果的缺陷，增强了测试码的鲁棒性．该算法对不同的连接权值都能得到满意的测试结果
|
| 关 键 词： | 神经网络 遗传算法 组合电路 测试生成 |
A Robust Algorithms for Test Generation Based on Neural Network |
| |
| Chen Chaoyang,Ding Mingyue.A Robust Algorithms for Test Generation Based on Neural Network[J].JOURNAL OF HUAZHONG UNIVERSITY OF SCIENCE AND TECHNOLOGY.NATURE SCIENCE,1999(9). |
| |
| Authors: | Chen Chaoyang Ding Mingyue |
| |
| Institution: | Chen Chaoyang Ding Mingyue |
| |
| Abstract: | Chakradhar et. al represented the combination circuit as a Hopfield neural network and exploited relaxation technique to generate the test vectors. But the connection weights among neurons have an effect upon the test result. A robust algorithms improved Genetic Algorithm is proposed, which is not sensitive to the connection weights. The experiments show that this algorithm is feasible for different connection weights. |
| |
| Keywords: | neural network genetic algorithm combination circuit test generation Chen Chaoyang Dr Institute of Pattern Recognition & AI HUST Wuhan 430074 China |
| 本文献已被 CNKI 等数据库收录！ |
