| High-Order BAM, Its Modifications and Performance Analyses |
| |
| 作者姓名： | 郑建立 陈松灿 |
| |
| 作者单位： | Zheng JianliCollege of Information Sciences & Technology，China Textile University，Shanghai，200051Chen SongcanDepartment of Computer Science，Nanjing University of Aeronaut. & Astronaut.，Nanjing，210000 |
| |
| 摘 要： | Based on Tai' s high - order bidirectional associative memory ( HOBAM) and Stmposon' s intraconnected BAM (IBAM) , two Improved models are first presented and discussed in this paper. The improved models not only retain the advantages of both HOBAM and ISAM but overcome the shortcomings of Kosko' s BAM also. Secondly their recall stabilities in synchronous and asyn-chronous update modes have been proven by defining corresponding energy functions which decrease as the re-call process proceeds such that the systems can ensure all the training pattern pairs to become local minima of the energy surfaces. Finally with signal - to - noise ratio (SNR) approach, we show that their storage capacities and error correction capabilities are better than that of the HOBAM.
|
High-Order BAM, Its Modifications and Performance Analyses |
| |
| Zheng JianliCollege of Information Sciences & Technology,China Textile University,Shanghai,Chen Songcan.High-Order BAM, Its Modifications and Performance Analyses[J].Journal of Donghua University,1998(4). |
| |
| Authors: | Zheng JianliCollege of Information Sciences & Technology China Textile University Shanghai Chen Songcan |
| |
| Institution: | Zheng JianliCollege of Information Sciences & Technology,China Textile University,Shanghai,200051Chen SongcanDepartment of Computer Science,Nanjing University of Aeronaut. & Astronaut.,Nanjing,210000 |
| |
| Abstract: | Based on Tai' s high - order bidirectional associative memory ( HOBAM) and Stmposon' s intraconnected BAM (IBAM) , two Improved models are first presented and discussed in this paper. The improved models not only retain the advantages of both HOBAM and ISAM but overcome the shortcomings of Kosko' s BAM also. Secondly their recall stabilities in synchronous and asyn-chronous update modes have been proven by defining corresponding energy functions which decrease as the re-call process proceeds such that the systems can ensure all the training pattern pairs to become local minima of the energy surfaces. Finally with signal - to - noise ratio (SNR) approach, we show that their storage capacities and error correction capabilities are better than that of the HOBAM. |
| |
| Keywords: | bidirectional associative memory ( BAM ) higher - order stability capacity SNR intraconnection neu-ral network |
| 本文献已被 CNKI 等数据库收录！ |
