双向联想记忆集成.pdfVIP

Dec．2007 Transactions of Nanjing University of Aeronautics＆Astronautics Vo1．24No．4 BIDIRECTIoNAL ASSoCIATIVE MEMoRY ENSEMBLE Wang M in ，Chu Rong (1．College of ComputerInformation Engineering，Hohai University，Nanjing，210098，P．R．China； 2．College of Electrical Engineering，Hohai University，Nanjing，210098，P．R．China) Abstract：The multiple classifier system (MCS)，composed of multiple diverse elassifters or feed—forward neura1 networks，can significantly improve the classification or generalization ability of a single classifier ． Enlightened bv the fundamental idea of MCS，the ensemble is introduced into the quick learning for bidirectiona1 associative mem— ory(QLBAM)to construct a BAM ensemble，for improving the storage capacity and the error—correction caDabil— ity without destroying the simple structure of the component BAM ． Simulations show that，with an appropriate overproduce and choose”strategy or thinning”algorithm，the proposed BAM ensemble significantly outper_ forms the single QLBAM in both storage capacity and noise—tolerance capability ． Key words：bidirectional associative memory；neural network ensemble；thinning algorithm CLC number：TP183 Document code：A Article ID：1005—1120(2007)04—0343—06 INTRoDUCTIoN tural change necessarily yields much more corn— plex connections and results in greater difficult