建立了神经网络的非平衡态热力学理论,给出了神经网络熵产生率,稳定条件及应用实例。
The non-equilibrium thermodynamic theory of artificial neural networks is introduced in this paper. The concepts of entropy production rate, network stability condition and its applications are given.
对具有密度依赖生育脉冲的模型,运用频闪映射,结合分支理论和数值分析,得到了正平衡态的存在性、稳定性;
For the model with time-dependent birth pulses, using the stroboscopic map, the theory on bifurcation and numerical analysis, the existence and stability of the positive equilibrium are obtained.
应用分歧理论及拓扑度理论的方法,得到了定态分歧解的存在性、唯一性及稳定性。
The results concerning the existence, the uniqueness and the stability of stationary bifurcation solutions have been obtained by the bifurcation theory and topological degree theory.
应用推荐