Computer Sciences

Algebraic Properties of Abstract Neural Network

Authors: Slepovichev I. I.

The modern level of neuroinformatics allows to use artificial neural networks for the solution of various applied problems. However many neural network methods put into practice have no strict formal mathematical substantiation, being heuristic algorithms. It imposes certain restrictions on development of neural network methods of the solution of problems. At the same time there is a wide class of mathematical models which are well studied within such disciplines as theory of abstract algebras, graph theory, automata theory.

Quantum Computers and Quantum Algorithms. Part 1. Quantum Computers

Authors: Solovyev V. M.

The paper presents the principles of operation of quantum computers. Competitive advantages of quantum computing are shown and some variants of a construction of an ideal quantum computer proposed. We analyze also the computational process in a quantum computer from the point of view of the complexity of algorithms. Implementation of nodes of a quantum computer is exemplified based on quantum communication schemes. The operation of Bloch sphere and visualization of the state of the qubit are described. Major obstacles to the creation of quantum computers are considered.

The Development of Software Components for Streaming Audio Content Filtering Through the Use of Hidden Markov Models

Authors: Savin A. N., Timofeeva N. E., Geraskin A. S., Mavlutova Y. A.

The results of the development of efficient algorithms for streaming voice recognition using stochastic models based on the use of hidden Markov models are shown in this work. The article provides basic theoretical information for the hidden Markov model of the discrete system and the necessary parameters to define it are distinguished. Also there are three main tasks considered that need to be solved for the successful application of hidden Markov models in speech recognition systems.

T-irreducible Extensions for Starlike Trees

Authors: Osipov D. Y.

We deal with a sort of optimal extensions of graphs, so called T-irreducible extensions. T-irreducible extension of a graph G is an extension of G obtained by removing a maximal set of edges from the trivial extension of G. A difficult starlike tree is a starlike tree that has at least one difficult node. T-irreducible extensions for nondifficult starlike trees were constructed by M. B. Abrosimov, T-irreducible extensions for palms (one of subclasses of starlike trees) were constructed by S. G. Kurnosova.

Reconstruction of a Labeled Graph by a Graph-walking Mobile Agent

Authors: Sapunov S. V.

The problem of construction of graph-like operational environment by a mobile agent is considered. The model of environment is defined as a simple undirected vertex labeled graph. We propose a polynomial time algorithm of graph reconstruction and labeling for the collective consisting of agent-explorer and agent-supervisor.

Parallelization of Parametric Synthesis by «Problems Portfolio» Scheme Based on MPI Technology

Authors: Andreichenko D. K., Eroftiev A. A., Melnichuk D. V.

We propose a prototype of a parallelization scheme «problems portfolio» by standard means of MPI technology with application to problems of parametric synthesis of controlled hybrid dynamical systems. Parametric synthesis of the stabilization system of a controlled mobile object was performed. Efficiency of the proposed algorithm of dynamical balancing of computational load is shown.

Principal Ideals in the Congruence Semilattice of a Path

Authors: Fomina E. O.

It is proven that principal ideals generated by congruences of a path having the same type are isomorphic lattices. The number of elements, atoms and coatoms is found for the principal ideal generated by a given congruence of a path.

The Application of a Genetic Algorithm to Global Optimization Problem Solving on Parallel and Distributed Computing Systems

Authors: Savin A. N., Druzhinin I. V., Eroftiev A. A.

This article presents the results of the adaptation of method of searching the global minimum of multiextremal criterion function of multiple variables with constraints based on genetic algorithm for parallel and distributed computing systems. Two variants of genetic algorithm parallelization are proposed. The reliability and performance of parallel versions of an algorithm, depending on its parameters and the number of nodes in parallel computer system is investigated.

On Varieties of Groupoids of Binary Relations

Authors: Bredikhin D. A.

In the paper, the bases of identities of varieties generated by classes of groupoids of the binary relations are found.

Abstract Characterization of Semigroups of Input Signals of Universal Planar Automata

Authors: Molchanov V. A.

Universal planar automata are universally attracted objects in the category of automata, for which sets of states and output signals are endowed with structures of planes. The main results of the paper give us necessary and sufficient conditions under which an arbitrary automaton is isomorphic to a universal planar automaton and an arbitrary semigroup is isomorphic to the semigroup of input signals of a universal planar automata.

 

Pages