|
«INFORMATICS AND APPLICATIONS»
Scientific journal
Volume 17, Issue 1, 2023
Content | About Authors
Abstract and Keywords
- I. N. Sinitsyn Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation, Moscow State Aviation Institute (National Research University), 4 Volokolamskoe Shosse, Moscow 125933, Russian Federation
Abstract: Exact and approximate analytical modeling methods for stochastic processes with invariant measure in Gaussian and non-Gaussian stochastic systems with unsolved derivatives are considered. The methods are based on the linear regression approximation of nonlinear functions with unsolved derivatives and reduction to stochastic Ito differential equations. Two exact methods for analytical modeling of one- and multidimensional distributions with invariant measure are described. Special attention is paid to normal approximation and parametrization methods. A test example for Duffing equation nonlinear in second derivative is given. The stationary and nonstationary regimes and asymptotic stability are investigated. The method of normal approximation for one- and two-dimensional distributions is accurate enough for engineering applications. Some generalizations concerning numerical analytical modeling are considered.
Keywords: analytical modeling; distribution parametrization; distribution with invariant measure; stochastic system; stochastic system with unsolved derivatives; stochastic process
- Yu. M. Kabanov M.V. Lomonosov Moscow State University, 1-52 Leninskie Gory, GSP-1, Moscow 119991, Russian Federation, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- A. P. Sidorenko M.V. Lomonosov Moscow State University, 1-52 Leninskie Gory, GSP-1, Moscow 119991, Russian Federation
Abstract: The authors study a model of clearing in an interbank network with crossholdings and default charges. Following the Eisenberg-Noe approach, the authors define the model via a set of natural financial regulations including those related to eventual default charges and derive a finite family of fixpoint problems. These problems are parameterized by vectors of binary variables. The model combines features of the Ararat-Meimanjanov, Rogers-Veraart, and Suzuki-Elsinger networks. The authors develop methods of computing the maximal and minimal clearing pairs using the mixed integer-linear programming and a Gaussian elimination algorithm.
Keywords: systemic risks; financial networks; clearing; crossholdings; default charges
- V. G. Ushakov Department of Mathematical Statistics, Faculty of Computational Mathematics and Cybernetics, M. V. Lomonosov Moscow State University, 1-52 Leninskie Gory, GSP-1, Moscow 119991, Russian Federation, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- N. G. Ushakov Institute of Microelectronics Technology and High-Purity Materials of the Russian Academy of Sciences, 6 Academician Osipyan Str., Chernogolovka, Moscow Region 142432, Russian Federation, Norwegian University of Science and Technology, 15A S. P. Andersensvei, Trondheim 7491, Norway
Abstract: Tests for normality are less sensitive to the data rounding than, for example, tests for exponentiality but among normality tests, the sensitivity is very different. In this paper, the authors find out which tests are more and which ones are less sensitive. The authors show that tests based on sample moments are much more robust with respect to the data rounding than tests based on order statistics (in contrast to the robustness with respect to outliers where order statistics are more robust than sample moments). This, however, only applies to the probability of Type I error. The probability of Type II error is very insensitive to the data rounding for all normality tests.
Keywords: normal distribution; test for normality; rounded data; significance level; Monte-Carlo simulation
- M. M. Leri Institute of Applied Mathematical Research of the Karelian Research Center of the Russian Academy of Sciences, 11 Pushkinskaya Str., Petrozavodsk 185910, Russian Federation
Abstract: In random configuration graphs with a discrete power-law vertex degree distribution with a fixed parameter, the average distance in the graph is considered, i. e., the arithmetic mean of distances between all pairs of graph nodes. This characteristic is estimated using simulation methods. Due to computational constraints, the author considers graphs in the pre-asymptotic domain (in this paper, these are the graphs up to 7000 nodes). The models of dependencies of the average distance on the graph size and the parameter of vertex degree distribution are reseived. The obtained results are compared with the results of theoretical studies of the typical distance in a graph in the asymptotics (i. e., when the number of graph vertices tends to infinity), given in the works by R. Hofstad
Keywords: configuration graph; power-law distribution; average distance in a graph; simulation
- R. G. Neychev Moscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny, Moscow Region 141700, Russian Federation
- I. A. Shibaev Moscow Institute of Physics and Technology, 9 Institutskiy Per., Dolgoprudny, Moscow Region 141700, Russian Federation
- V. V. Strijov Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: The paper investigates the problem of regression model generation. A model is a superposition of primitive functions. The model structure is described by a weighted colored graph. Each graph vertex corresponds to a primitive function. An edge assigns a superposition of two functions. The weight of an edge is equal to the probability of superposition. To generate an optimal model, one has to reconstruct its structure from its graph adjacency matrix. The proposed algorithm reconstructs the minimum spanning tree from the weighted colored graph. The paper presents a novel solution based on the prize-collecting Steiner tree algorithm. This algorithm is compared with its alternatives.
Keywords: symbolic regression; linear programming; superposition; minimum spanning tree; adjacency matrix
- A. A. Grusho Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- N. A. Grusho Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- M. I. Zabezhailo Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- V. V. Kulchenkov VTB Bank, 43-1 Vorontsovskaya Str., Moscow 109147, Russian Federation
- E. E. Timonina Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- S. Ya. Shorgin Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: In the present paper, a classification object is considered as the cause for the appearance of one or more consequences and any classification algorithm decides on the class observing the consequences from the analyzed cause. The paper considers the consequences of the cause in the binary classification problem as sources of additional information confirming or rejecting the hypothesis of the cause in the classified object. When considering a hypothesis about the presence or absence of a certain cause in an object classified by this property, the knowledge presentation language is automatically built based on several consequences. Then, it is easy to use the available information from different information spaces in an object classification task. To use cause-and-effect relationships in a classification task, machine learning should be used. In conditions of teaching with a teacher, there are many precedents when the presence of a cause is known. Then one can statistically single out events that are the consequences of the cause. Deterministic cause-and-effect relationships generate errors only at the expense of noise. In those precedents where there is no cause, positive classification appears only at the expense of noise regardless of precedent to precedent. Thus, even a weak deviation from equally probable noise allows one to build a consistent criterion that distinguishes consequences from random noise. Sequelae can be isolated independently of each other. This follows from the determinism of the cause-and-effect relationship and the independence of noise.
Keywords: finite classification task; cause-and-effect relationships; machine learning
- P. O. Arkhipov Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- S. L. Philippskih Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- M. V. Tsukanov Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: A new model ofa stepped convolutional neural network for classifying anomalies in panoramas has been developed. Appropriate datasets for classification are selected. The conclusion is made about the incompleteness of the method previously used by the authors to find anomalies in special areas with high color difference in panoramas. The search for these areas by the previously developed method did not set the task of their classification.
For automatic identification of detected objects, it is proposed to apply deep learning models using suitable neural networks. Particular attention is paid to work with data containing unbalanced classes and images of different sizes. The results of image classification of popular architectures of neural networks are compared with the newly developed stepped convolutional neural network.
Keywords: panoramic image; data set; multilabel classification; stepwise convolutional neural network; ensemble; transfer learning
- I. N. Rozenberg Research & Design Institute for Information Technology, Signalling and Telecommunications on Railway Transport, 27-1 Nizhegorodskaya Str., Moscow 109029, Russian Federation
- S. K. Dulin Research & Design Institute for Information Technology, Signalling and Telecommunications on Railway Transport, 27-1 Nizhegorodskaya Str., Moscow 109029, Russian Federation, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- N. G. Dulina Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: The initial syntactic level of interoperability involves communication with the appropriate protocol, hardware, software, and necessary level of data compatibility. The work is devoted to the study of the level of compatibility of data describing interacting elements based on the feature vector. To do this, a model of structural correspondence is proposed which allows assessing the tendency to establish interoperability. Modeling structural interoperability based on the analysis of signs of connections using the introduced consistency criterion leads to finding the closest consonant pre-image to the original set. The found consonant pre-image with its subsets indicates the preferred grouping of elements in which the interoperability between them is established with the least mismatch with respect to the fixed signs of connections. Since the elements under consideration are described by a vector of parameters, from the comparison of which one can infer the similarity between the elements, respectively, the presence of elements in the same subset indicates a potential motivation for interoperability
Keywords: interoperability; structural consistency; connectivity matrix
- S. V. Listopad Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- I. A. Kirikov Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: The paper continues research on computer simulation with hybrid intelligent multiagent systems of a teamwork of specialists of various profiles who solve problems at a round table. The agents of such systems are autonomous software entities that imitate the reasoning of real specialists. Modeling of heterogeneous knowledge, goals, and points of view of agents on the problem posed within single intelligent system causes their collision, the emergence of conflicts by analogy with how it happens in simulated teams. Not every conflict between agents is destructive and requires suppression: conflict management in a hybrid intelligent multiagent system as well as in a team involves the identification of a decision-making situation, if necessary, stimulation and subsequent resolution of constructive forms of conflict as well as the prevention of its destructive forms. The paper proposes the method based on fuzzy rules to manage conflicts between agents in hybrid intelligent multiagent systems.
Keywords: conflict; hybrid intelligent multiagent system; team of specialists; conflict management
- A. V. Bosov Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation, Moscow State Aviation Institute (National Research University), 4 Volokolamskoe Shosse, Moscow 125933, Russian Federation
- A. N. Ignatov Moscow State Aviation Institute (National Research University), 4 Volokolamskoe Shosse, Moscow 125933, Russian Federation
Abstract: The problem of assessing and analyzing traffic accidents risk on the rail transport is considered. Two functions of the integral risk are proposed that allow assessing danger of transportation along the entire route of a transport. The probability of an unfavorable event occurring during transportation and the expected damage are chosen as such functions. The concept of assessing probability and damage from unfavorable events during the freight trains transportation is proposed. A meaningful example of calculating integral risk functions is given on the basis of previously investigated statistics on the freight trains transportation and unfavorable events that occurred with them.
Keywords: risk; unfavorable event; rail transport; probability; expected damage
- Yu. E. Malashenko Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- I. A. Nazarova Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: A method for estimating resources with an equalizing distribution of internodal loads in a multiuser
network is proposed. Within the framework of a formal mathematical model, the capacity of edges is considered as
components of a vector of resources that are required for the transmission of different types of flows. An algorithmic
procedure for the redistribution and usage of capacity with equal quota of resources for all pairs is proposed. When
searching for the corresponding edge loads, the values of the maximum single-product flows for each pair of nodes
are determined. In the course of computational experiments, the total resource is considered to be set for networks
with various structural features.
Keywords: multicommodity flow model; network resource distribution and internodal loads; network peak load
- Ya. M. Agalarov Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: The problem of optimal control of an additional server in a stationary G/M/s queue is considered.
The additional server can be turned on and off at instants when the queue length is changed. It is formulated as the nonlinear optimization problem, in which the objective function accounts for amounts for service, losses due to the waiting of customers, maintenance, and downtime of the additional server. The functioning of the system is described as a controlled Markov chain. Only stationary control policies are considered. For Poisson arrivals, necessary and sufficient conditions are given for the existence of the optimal decision point (threshold) and it is proved that the objective function is unimodal. A simple algorithm for the computation of the threshold is provided.
Keywords: multiserver queuing system; optimization; additional server
- K. Y. B. Adou Peoples' Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation
- E. V. Markova Peoples' Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation
- Yu. V. Gaidamaka Peoples' Friendship University of Russia (RUDN University), 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation, Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
- S. Ya. Shorgin Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: The network slicing (NS) technology, which has been actively studied in recent years, is based on the representation of a common network infrastructure in the form of various customizable logical networks called slices and involves the division of mobile network operators into two groups - physical network infrastructure providers (InPs) and mobile virtual network operators (MVNOs). The MVNOs lease the physical resources of InPs to create their own slices to provide services to their users with different quality of service requirements. In the
present paper, for a network with NS technology, a scheme for accessing its radio resources is proposed that provides
users with services with a guaranteed bit rate (GBR) and priority control based on the implementation of the user
service interruption mechanism. The authors propose a scheme for accessing radio resources of a network under
NS technology that provides users with services with GBR and priority control based on the implementation of the
user service interruption mechanism. To evaluate the effectiveness of the proposed scheme, a comparative analysis
of its characteristics with the characteristics of the access scheme based on the resource reservation mechanism was
carried out.
Keywords: 5G; network slicing; quality of service; key performance indicators; priority management; service interruption; iterative method
- G. A. Agasandyan Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: The work continues studying problems of using continuous VaR-criterion (CC-VaR) in financial markets.
Again some technical problems are concerned. However, they emerge this time not in multidimensional relatively simple binary markets but in multidimensional markets that are an extension of one-dimensional traditional markets of options such as calls and puts. In assumption that scenario butterflies are not traded in markets directly a method of receiving their replication from multidimensional options, i. e., а-options, is developed. It is based on options parity theorems and can be applied to markets of arbitrary dimension, but actual realization is conducted for two-dimensional markets. The bases constructions in terms of а-options both one-type and natural mixed with
selected market center are produced. Theoretical representations of optimal portfolios in these bases accompanied with the payoffs diagram are illustrated by the distinctive example of a two-dimensional market.
Keywords: underliers; multidimensional market; investor's risk preferences function; continuous VaR-criterion; cost and forecast densities; scenario indicators; bases; binary options; one-type portfolio; market center; mixed portfolio
- I. M. Zatsman Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
Abstract: Three basic notions of informatics - data, information, and knowledge - are considered. The variant of specification of these notions within the framework of constructing a system of terms of the scientific paradigm of informatics as a fundamental science is proposed. On the one hand, the notions of "data," "information," and "knowledge" are widely used in the scientific literature and textbooks on informatics, in particular, when describing its theoretical foundations. On the other hand, there is still no consensus on their semantic content.
The current situation is most likely due to the widespread presupposition (implicit assumption) that the notions in question express some objective entities of the subject domain of informatics. The paper assumes that they express intersubjective entities that by their nature arise as objects of thought during an agreement process, that is, they are not objectively existing. From the point of view of the Frege's triangle (subject - notion (concept) - word that expres the notion and denote the subject), for objective entities, the primary vertex of the triangle is the subject as a result of the study of which the notion and the word appear. For intersubjective entities, the primary vertex is the notion, the definitions of which must be discussed in the interests of reaching consensus. If it can be achieved, then it is during the process of discussion that the subject of thought appears, the word denoting it and expressing the notion which together form the Frege's triangle. The aim of the paper is to specify the basic notions of informatics as expressing intersubjective entities and being the primary vertices of the Frege's triangle, to distribute them among the media of the subject domain of informatics, highlighting the boundaries between the media, and to consider the relationship between the specified notions on these boundaries.
Keywords: scientific paradigm; informatics as fundamental science; data; information; knowledge; intersubjective entities of informatics; Frege's triangle
|
