后向传播的英文

• back-propagating

• "后"英文翻译    behind; back; rear
• "向"英文翻译    direction
• "传播"英文翻译    disseminate; propagate; spre ...
• "后向传播，反向传播" 英文翻译 :    back propagation
• "定时后向传播" 英文翻译 :    btt backpropagation through time
• "误差后向传播" 英文翻译 :    ebp
• "定向传播" 英文翻译 :    beaming; guided propagation
• "反向传播" 英文翻译 :    b back propagation; bback propagation; counterpropagation
• "双向传播" 英文翻译 :    bidirectional propagation; two-way communications
• "正向传播" 英文翻译 :    forward-propagating
• "准横向传播" 英文翻译 :    quasi-transverse
• "单向传播时间" 英文翻译 :    one-way propagation time
• "对向传播网络" 英文翻译 :    counter propagation network
• "反向传播光子" 英文翻译 :    oppositely-directed photons
• "反向传播路径" 英文翻译 :    reverse travel path
• "反向传播算法" 英文翻译 :    bpa back propagation algorithm
• "横向传播常数" 英文翻译 :    transverse redundancy check
• "误差反向传播" 英文翻译 :    ebp
• "正向传播路径" 英文翻译 :    forward travel path
• "轴向传播常数" 英文翻译 :    axial propagation constant
• "反方向传播的光子" 英文翻译 :    oppositely directed photons
• "反方向传播光子" 英文翻译 :    oppositely-directed photons
• "水平传送 横向传播" 英文翻译 :    horizontaltransmission
• "自联想反向传播" 英文翻译 :    abp
• "地向传播等离子体团" 英文翻译 :    earthward propagating plasmoids

• Changes to any phase can be propagated forward or back to keep the model consistent , and differences are reported
  任何阶段的改变可以前向或后向传播以保持模型的连续性，并报告差别。
• Based on analyzing the multilayer perceptron neural network , it induces the error backpropagation algorithm while taking the hyperboloid tangential function as nonlinear activation function
  在分析多层感知器神经网络的基础上，推导了双曲正切函数为激励函数的误差后向传播算法。
• Aiming at the disadvantages of present feature association algorithm , the modified bp neutral network is applied to feature association and has a higher association rates
  4针对现行属性关联算法的缺点，把改进的误差后向传播( bp )神经网络方法应用于属性关联，该方法有较高的关联成功率。
• The paper mainly did the dissemination of error neural network ( bp ) research and gave recommendations on ways to improve bp algorithm . by using better diagnostic techniques with the neural networks legend , memory and reasoning functions and tolerant nature , robustness and good nonlinear , it can better realized the fault
  主要对误差后向传播神经网络( bp )进行研究,提出了改进的bp算法,利用神经网络的联想、记忆和推理功能以及容错性、鲁棒性和很好的非线性映射能力等特点,更好地实现故障诊断。
• First is the integration of back propagation , clustering and decision tree technology in a seamless hole . second is the improvement of three - layer architecture . and weights are detailed to the connection between different feature - value pairs and cases instead of to the connection between features and cases
  一是在同一模型中融合了后向传播算法、聚类技术、决策树技术；二是把实例的三层结构进行了改进，并把实例与属性之间的连接权重细化到实例与属性取值间的联接权重；三是在后向传播算法中引入差分hebb学习法，并在计算过程中充分考虑了历史数据的影响。