简介：Thecapabilityofmultilayerperceptrons(MLPs)forapproximatingcontinuousfunctionswitharbitraryaccuracyhasbeendemonstratedinthepastdecades.Backpropagation(BP)algorithmisthemostpopularlearningalgorithmfortrainingofMLPs.Inthispaper,asimpleiterationformulaisusedtoselecttheleamingrateforeachcycleoftrainingprocedure,andaconvergenceresultispresentedfortheBPalgo-rithmfortrainingMLPwithahiddenlayerandalinearoutputunit.Themonotonicityoftheerrorfunctionisalsoguaranteedduringthetrainingiteration.

简介：Letfibeaboundedsmoothdomain.Inthispaper,theauthorsdefinetheBesovspacesBpa,pon,establishtheatomicdecompositionofthesespaces,andobtaintheregularityestimateoftheDirichletproblemandtheNeumannproblemfortheLaplaceoperatoronthesespaces.

简介：文章分析了传统BP学习方法的缺陷，给出了一种改进的学习方法，并用非线性函数tg△x和(e^△x-1)代替传统的线性函数△x进行网络学习和参数调整．仿真表明，该算法能有效克服网络陷入局部极小的困境，并大大提高收敛速度．

简介：本文利用BP神经网络建立起66例肝硬化治疗结果数据预测模型，并基于matlab得出预测结果。实验证明利用BP神经网络可有效地预测肝病治疗效果。