登入選單

返回Google圖書搜尋

Decision Theoretic Generalizations of the PAC Learning Model

David Haussler

出版	University of California, Santa Cruz, Computer Research Laboratory, 1990
URL	http://books.google.com.hk/books?id=HfwfAQAAIAAJ&hl=&source=gbs_api

註釋

Abstract: "We describe a generalization of the PAC learning model that is based on methods from statistical decision theory. This generalization allows us to give a unified treatment of the problem of learning a function from a set X into an arbitrary set Y, the problem of learning a conditional probability distribution on Y given X (called a p- concept when Y=[0,1]), and the problem of learning a distribution on the set X itself (i.e. density estimation). The key changes in the PAC model are to let the examples be generated by an arbitrary joint distribution on X x Y, and to use a loss function to define the error of a hypothesis