Learning How to Separate
نویسندگان
چکیده
The main question addressed in the present work is how to find effectively a recursive function separating two sets drawn arbitrarily from a given collection of disjoint sets. In particular, it is investigated in which cases it is possible to satisfy the following additional constraints: confidence where the learner converges on all data-sequences; conservativeness where the learner abandons only definitely wrong hypotheses; consistency where also every intermediate hypothesis is consistent with the data seen so far; set-driven learners whose hypotheses are independent of the order and the number of repetitions of the dataitems supplied; learners where either the last or even all hypotheses are programs of total recursive functions. The present work gives an overview of the relations between these notions and succeeds to answer many questions by finding ways to carry over the corresponding results from other scenarios within inductive inference. Nevertheless, the relations between conservativeness and set-driven inference needed a novel approach which enabled to show the following two major results: (1) There is a class for which recursive separators can be found in a confident and set-driven way, but no conservative learner finds a (not necessarily total) separator for this class. (2) There is a class for which recursive separators can be found in a confident and conservative way, but no set-driven learner finds a (not necessarily total) separator for this class. Supported in part by NUS grant number RP3992710. Supported by the Heisenberg grant Ste 967/1–1 of the Deutsche Forschungsgemeinschaft (DFG).
منابع مشابه
Blind Signal Separation Using an Extended Infomax Algorithm
The Infomax algorithm is a popular method in blind source separation problem. In this article an extension of the Infomax algorithm is proposed that is able to separate mixed signals with any sub- or super-Gaussian distributions. This ability is the results of using two different nonlinear functions and new coefficients in the learning rule. In this paper we show how we can use the distribution...
متن کاملBlind Signal Separation Using an Extended Infomax Algorithm
The Infomax algorithm is a popular method in blind source separation problem. In this article an extension of the Infomax algorithm is proposed that is able to separate mixed signals with any sub- or super-Gaussian distributions. This ability is the results of using two different nonlinear functions and new coefficients in the learning rule. In this paper we show how we can use the distribution...
متن کاملHow to Train in Sports Clubs in Tehran: Nonlinear Versus Linear Method
Nonlinear methods in physical education is a new idea that covers all the disadvantages of the traditional method and is effective for learning and implementation based on the results of studies compared to the linear method. However, it is not clear how much this method is used by trainers in Iran. This study sought to determine what kind of method (linear / non-linear) instructors use for tra...
متن کاملThe Effects of Separate and Combined Stress during Pregnancy on Motor Learning of Offspring of Rats
Background & Aims: Today, due to the modern industrial life of human beings, stress has become prevalent among them and they will suffer from its complications. Exposure to stress during pregnancy can change many babies' normal physiological processes. The separate and combined effects of three common types of prenatal stress were investigated on motor learning of male offspring of rats. Method...
متن کاملMotor Learning and Movement Performance: Older versus Younger Adults
Introduction: Motor skills play an important role during life span, and older adults need to learn or relearn these skills. The purpose of this study was to investigate how aging affects induction of improved movement performance by motor training. Methods: Serial Reaction Time Test (SRTT) was used to assess movement performance during 8 blocks of motor training. Participants were tested i...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
برای دانلود متن کامل این مقاله و بیش از 32 میلیون مقاله دیگر ابتدا ثبت نام کنید
ثبت ناماگر عضو سایت هستید لطفا وارد حساب کاربری خود شوید
ورودعنوان ژورنال:
- Theor. Comput. Sci.
دوره 313 شماره
صفحات -
تاریخ انتشار 2001