| STABILITY OF THE HIGH-ORDER BAM AND ITS PERFORMANCE ANALYSIS |
| |
| 引用本文： | 陈松灿. STABILITY OF THE HIGH-ORDER BAM AND ITS PERFORMANCE ANALYSIS[J]. 南京航空航天大学学报(英文版), 1997, 0(1) |
| |
| 作者姓名： | 陈松灿 |
| |
| 作者单位： | 南京航空航天大学计算机科学与工程系 |
| |
| 摘 要： | 由Ｔａｉ提出的高阶双向联想存储器（ＨＯＢＡＭ）推广了Ｋｏｓｋｏ的ＢＡＭ，它不仅拥有ＢＡＭ的优点，而且放宽了连续性假定以得到可靠的回忆，并大大改进了ＢＡＭ的存储容量和纠错性能。然而，Ｔａｉ对于ＨＯＢＡＭ的性能分析仅限于计算机模拟，并没有给出理论分析结果。本文填写了此方面空白，并给出ＨＯＢＡＭ稳定性和存储容量的理论证明，使得该系统能从理论上得以保证所有训练模式对成为其稳定点。
|
| 关 键 词： | 神经网络；联想存储器；稳定性；存储容量 |
STABILITY OF THE HIGH ORDER BAM AND ITS PERFORMANCE ANALYSIS |
| |
| Chen Songcan Zhu Wujia. STABILITY OF THE HIGH ORDER BAM AND ITS PERFORMANCE ANALYSIS[J]. Transactions of Nanjing University of Aeronautics & Astronautics, 1997, 0(1) |
| |
| Authors: | Chen Songcan Zhu Wujia |
| |
| Affiliation: | Chen Songcan Zhu Wujia Department of Computer Science and Engineering,NUAA29 Yudao Street,Nanjing 210016,P. R. China |
| |
| Abstract: | |
| |
| Keywords: | neural networks associative memories stability storage capacity |
| 本文献已被 CNKI 等数据库收录！ |
