''It is true that this theory came from the same roots and uses the same formal tools as statistics. However, to belong to the statistical branch of science this is not enough. Much more important is to share the same belief in the models and to share the same philosophy. Whatever you are sugges-ting is not in the spirit of what I am doing or what A. Kolmogorov is doing. It is not what our students are doing nor will it be what the students of our students do. Therefore, you must have your own students, develop your own philosophy, and create your own community.'' [Gnedenco in response to Vapnik]
''In the early 1930s Popper suggested his idea of falsifiability. It was considered both as the demarcation line between metaphysics and natural science as well as a justification of Occam’s Razor principle. Popper developed the falsifiability idea over his entire lifetime: his first publication appeared in German in 1934 his last addition was in an English edition in 1972. This idea is considered as one of the most important achievements in the philosophy of science of the 20th century. Fisher’s philosophy of applied statistics and Popper’s justification of the dependence of generalization ability on the number of entities formed the classical paradigm of philosophical realism for induction. The continuation of the Kolmogorov-Glivenco-Cantelli line of theoretical statistics led to the development of the VC theory that reflected the philosophical instrumentalism paradigm'' [Vapnik]
''In the early 1930s Popper suggested his idea of falsifiability. It was considered both as the demarcation line between metaphysics and natural science as well as a justification of Occam’s Razor principle. Popper developed the falsifiability idea over his entire lifetime: his first publication appeared in German in 1934 his last addition was in an English edition in 1972. This idea is considered as one of the most important achievements in the philosophy of science of the 20th century. Fisher’s philosophy of applied statistics and Popper’s justification of the dependence of generalization ability on the number of entities formed the classical paradigm of philosophical realism for induction. The continuation of the Kolmogorov-Glivenco-Cantelli line of theoretical statistics led to the development of the VC theory that reflected the philosophical instrumentalism paradigm'' [Vapnik]
|
|