neural network training algorithm based on particle swarm optimization 神經(jīng)網(wǎng)絡(luò)基于粒子群優(yōu)化的學(xué)習(xí)算法研究
2 . standard samples that are needed for neural network training are made 2.制作神經(jīng)網(wǎng)絡(luò)訓(xùn)練所需的標(biāo)準(zhǔn)樣本。
neural network training for system identification using a least square method 神經(jīng)網(wǎng)絡(luò)用于系統(tǒng)辨識(shí)的一種最小二乘法
neural network trained with particle swarm algorithm and its application to nonlinear system identification 粒子群神經(jīng)網(wǎng)絡(luò)及其在非線性系統(tǒng)辨識(shí)中的應(yīng)用
The neural network training algorithm is improved, which is based on instantaneous optimal control method, taking into consideration of the energy of earthquake 摘要改進(jìn)一種基于瞬時(shí)最優(yōu)控制的神經(jīng)網(wǎng)絡(luò)訓(xùn)練算法。
The application of hybrid algorithm which combines improved genetic algorithm and error back-propagation algorithm in artificial neural network training is studied first 首先研究了將改進(jìn)遺傳算法和誤差反向傳播(bp)算法相結(jié)合的混合算法來訓(xùn)練人工神經(jīng)網(wǎng)絡(luò)。
The advanced and efficient algorithm variable scale method for learning is used in bp neural network training . finally the control simulation results are given 在bp網(wǎng)絡(luò)的訓(xùn)練過程中,采用了自調(diào)整的學(xué)習(xí)算子以加速收斂得到較好的學(xué)習(xí)效果,最后給出了仿真結(jié)果。
Finally, in order to solve the problem of getting the sample of input / output, a neural networks training algorithm is proposed that is based on instantaneous optimal control method 針對(duì)訓(xùn)練樣本對(duì)難以獲取的問題,提出了基于瞬時(shí)最優(yōu)控制神經(jīng)網(wǎng)絡(luò)的建筑結(jié)構(gòu)主動(dòng)控制。
The fault diagnosis example shows that the difficulty of neural network training is diminished and the fault diagnosis accuracy can reach more then 99 % when faults overlaps exist 通過診斷示例表明,該方法在故障類存在重疊時(shí),降低了神經(jīng)網(wǎng)絡(luò)的訓(xùn)練難度,故障診斷的正確率達(dá)到99%以上。