Two problems concerning solid variable function 关于实变函数论的两个问题
To investigate generalization capability of feedforward neural networks ; the influencing factors of generalization capability of feedforward neural networks are analyzed according to function theory 摘要针对前向神经网络泛化问题,从函数论的角度分析了影响前向神经网络泛化性能的因素。
Therefore , computation of the bergman kernel function by explicit formula is an important research direction in several complex variables . up to now , there are still many mathematicians working in this direction 因此,如何求bergman核函数的显表达式一直是多复变函数论的一个重要的研究方向,至今仍吸引着许多数学家对此进行研究。
In the fifth chapter , we use the the methods of function theory establised a class of analytic reproducing kernel space , we call it a generalized arveson space . we study the relationship of these space , and obtain some intersting results , we also defined toeplitz c * - algebra on the generalized areveson space , obtain some results of toeplitz c * - algebra on the generalized areveson space 在第五章中,我们利用函数论的方法建立了更广泛的一类解析的再生核空间,称之为广义arveson空间,讨论了它的一些关系,并得到了一些有趣的结果。还定义了广义arveson空间上的toeplitzc * -代数,建立了广义arveson空间上的toeplitzc * -代数的一些性质。