二是“试验机制无差别原则”,用以为主观主义概率归纳逻辑的意见收敛定理进行辩护。
The other is the Indifference Principle of Trial Setup, with which the Theorem of Opinions' Convergence in the subjective probability theory is justified.
针对施工数据,根据数据归一化的原则,确定了一个比较合理的归一化方法,提高了网络学习的效率和网络收敛的速度。
A reasonable normalization method is made certain by the normalization principle and the data of construction. This method increases the learning efficiency and the rate of convergence of the network.
并且详细叙述了神经网络结构参数如隐含层神经元个数、激励函数、网络收敛精度等的确定原则和方法。
The principle and methods to determine the network parameters such as number of neuron in hidden layer, excitation function and the convergence accuracy have been analyzed in detail.
应用推荐