scholarly journals SELF-ORGANIZING AND SELF-REPAIRING MASS MEMORIES FOR AUTOSOPHY MULTIMEDIA ARCHIVING SYSTEMS - Replacing the Data Processing Computer with Self-Learning Machines based on the Autosophy Information Theory

2021 ◽  
Vol 22 (4) ◽  
pp. 171-180
Author(s):  
V. B. Melekhin ◽  
M. V. Khachumov

We formulate the basic principles of constructing a sign-signal control for the expedient behavior of autonomous intelligent agents in a priori undescribed conditions of a problematic environment. We clarify the concept of a self-organizing autonomous intelligent agent as a system capable of automatic goal-setting when a certain type of conditional and unconditional signal — signs appears in a problem environment. The procedures for planning the expedient behavior of autonomous intelligent agents have been developed, that imitate trial actions under uncertainty in the process of studying the regularities of transforming situations in a problem environment, which allows avoiding environmental changes in the process of self-learning that are not related to the achievement of a given goal. Boundary estimates of the proposed procedures complexity for planning expedient behavior are determined, confirming the possibility of their effective implementation on the on-board computer of the automatic control system for the expedient activity of autonomous intelligent agents. We carry out an imitation on a personal computer of the proposed procedures for planning purposeful behavior, confirming the effectiveness of their use to build intelligent problem solvers for autonomous intelligent agents in order to endow them with the ability to adapt to a priori undescribed operating conditions. The main types of connections between various conditional and unconditional signal — signs of a problem environment are structured, which allows autonomous intelligent agents to adapt to complex a priori undescribed and unstable conditions of functioning.


Author(s):  
Yevgeniy Bodyanskiy ◽  
Valentyna Volkova ◽  
Mark Skuratov

Matrix Neuro-Fuzzy Self-Organizing Clustering NetworkIn this article the problem of clustering massive data sets, which are represented in the matrix form, is considered. The article represents the 2-D self-organizing Kohonen map and its self-learning algorithms based on the winner-take-all (WTA) and winner-take-more (WTM) rules with Gaussian and Epanechnikov functions as the fuzzy membership functions, and without the winner. The fuzzy inference for processing data with overlapping classes in a neural network is introduced. It allows one to estimate membership levels for every sample to every class. This network is the generalization of a vector neuro- and neuro-fuzzy Kohonen network and allows for data processing as they are fed in the on-line mode.


Author(s):  
Сергей Юдин ◽  
Sergey Yudin ◽  
Александр Юдин ◽  
Aleksandr Yudin

The monograph contains a number of new author's theoretical and methodological developments based on the methods of mathematical information theory. It can be useful for students, graduate students and researchers in the analysis of data and the construction of models of economic, sociological and psychometric processes. It can be used as a textbook in the study of the relevant sections of the course "Mathematics", "modeling And data processing", "Econometrics" and others.


Kybernetes ◽  
1998 ◽  
Vol 27 (3) ◽  
pp. 264-272 ◽  
Author(s):  
Rallis C. Papademetriou

Sign in / Sign up

Export Citation Format

Share Document