The Continuous Newton's Method, Inverse Functions, and Nash-Moser
The Continuous Newton's Method, Inverse Functions, and Nash-Moser
复制标题
连续牛顿法、反函数和 Nash-Moser
DOI:
10.1080/00029890.2007.11920431
复制
发表时间:
2007
期刊:
影响因子:
--
通讯作者:
J. Neuberger
中科院分区:
文献类型:
--
作者:
E. Scheinerman;J. Neuberger
hoping that z\, z2, ... converges to a zero of F. What can stop this process from find ing a zero of F? For one thing, there might not be a zero of F. For another, the process might terminate for some integer k in the event that F'izk) does not have an inverse. A domain of attraction corresponding to a given root of F consists of the set of all starting values z0 that lead, through convergence of z\, z2, ..., to this root. Newton's method can lead to chaotic domains of attraction, even for simple choices of F (see [8]). This can lead to striking pictures but constitutes a nightmare for the numerical analyst. This fact, if nothing else, leads one to the damped Newton's method, which consists of