抄録
Offer Organization: Japan Society for the Promotion of Science, System Name: Grants-in-Aid for Scientific Research, Category: Grant-in-Aid for Scientific Research (C), Fund Type: competitive_research_funding, Overall Grant Amount: - (direct: 3600000, indirect: 1080000)
We succeed an improvement of the dynamical scaling by means of the kernel method, by which a scaling function can be expressed by the kernel functions avoiding some trial and error processes. An efficient and fast estimation method is realized by means of the conjugate gradient method for optimization. The process for scaling is almost automatized providing various applications. The bootstrap method by the use of huge number of data points provides a remarkable improvement for the accuracy of estimations. Comparing dynamical scaling methods with assuming several patterns for asymptotic forms of relaxation time and that without any assumption for the form, we develop a method to discriminate the transition types. We also establish the reliable analysis for topological phase transitions such as the KT one.