The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
We have shown that, in some sense, computers can be taught how to learn how to learn. The mathematical result constructed sequences of functions that were easy to learn, provided they were learned one at a time in a specific order. Furthermore, the sequences of functions constructed above are impossible to learn, by an algorithmic device, if the functions are not presented in the specified order....
A logic procedure is provided as to discover relations inductively from experimental data. This logic inductive inference method discovers the properties of the relations in question step by step in the form of fixed point equations. The defining formulas in these equations may use only bounded quantifiers but the relation to be defined may also negatively occur. By using this type of equations...
This paper treats the inductive inference of computable functions the observations of which are falsified by noise. The effect of noise is assumed to satisfy a recursion theoretic randomness condition. It turns out that under three natural assumptions (finite range, reliable identifiability of the function class and "proper" noise function) the identifiability is preserved up to a finite...
We present in this paper a formal theory of analogical reasoning. We are mainly concerned with three subjects: a formal definition of analogy, a formalization of the reasoning in terms of deduction, and a method for realizing the reasoning in a logic programming system. First we assume that each domain for the reasoning is the least model for logic program. Then we consider an analogy as a partial...
The inductive reasoning core of a medical expert system is presented. The goal is to generate a knowledge base containing diagnostic rules from an, in some sense incomplete, data base. An algorithm is described which, using the missing information in a special way, generates the required rules. The generated rules also match an appropriate complete data base that is exponentially greater than the...
The present paper is motivated by the aim to use inductive inference methods in the field of automatic program synthesis. For that reason, we give an algebraic framework for a comprehensive generalization of SUMMERS' approach by using Church-Rosser specifications of a special kind. Based on a brief description of a general program synthesis method we deduce a decidability problem forming a pure theoretical...
Before we describe our approach to the problem of learning the action part of IF(pattern) THEN-DO(action)-rules we give a survey to the problem of knowledge acquisition for expert systems. In connection with our work we focus on automatic knowledge acquisition by learning methods.
In this paper first we study the enumeration technique, as essentially the only method of the inductive inference, comparing it with best strategies. We conclude that the original enumeration strategy doesn't usually from experience: it is controlled by mistakes or of negative feedback. We define a special subclass of enumeration strategies that is called asymptotic ones. It seems that such strategies...
Synthesis of programs of recursive functions from input/output examples is investigated where the synthesis is effective in the sense that one is able to algorithmically determine when the correct program of the function under consideration has been found. Necessary and sufficient conditions are derived that the synthesis can be done if the admissible number of input/output examples is bounded.
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.