«INFORMATICS AND APPLICATIONS»
Scientific journal
Volume 18, Issue 2, 2024

Content | About  Authors

Abstract and Keywords

ON FUNCTOR REPRESENTATION OF OPTIMIZED DYNAMIC MULTIAGENT SYSTEMS

• N. S. Vasilyev  N. E. Bauman Moscow State Technical University, 5-1 Baumanskaya 2nd Str., Moscow 105005, Russian Federation

Abstract: Functors' topoi is chosen as a computational tool for synthesizing dynamic multiagent systems (DMAS). The scale orders the objects as multiagent system states to solve attendant static subgames in them. The initial dynamic game and all static subproblems are represented in the monoidal category of binary relations. Players' preference relations might be maximized in DMAS. The game rational solution is understood as equilibrium. The compositional structure of the optimized DMAS can be described in the form of the game dynamic resulting relation (DRR). Players' rational behavior search is reduced to DRR subsequent maximization. For this purpose, the Bellman's method is generalized to solve control problems stated in the form of relations. The program implementation of the approach can be based on neural networks due to the consistency of the architectures of the applied relation graphs and neural networks.

Keywords: functor category; compositionality; monoidal category; opposite image; game dynamic relation; static subgame; preference relation; dynamic resulting relation; rational solution; Bellman morphism

MARKET WITH MARKOV JUMP VOLATILITY V: MARKET COMPLETION WITH DERIVATIVES

• A. V. Borisov  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: The final, fifth, part of the series is devoted to a replenishment procedure of the market with a Markov jump regime change. The market includes a riskless bank deposit with a known nonrandom interest rate and a set of underlying risky assets. The instantaneous interest rates and volatilities of the assets are the functions of the hidden regime change factor described by some finite state Markov jump process. The purpose of this article is to complete the investigated market. It means the market enlargement by a set of auxiliary financial instruments. The point is that any contingent claim declared in the market can be replicated with some self-financing portfolio containing the original and auxiliary instruments. For the market completion, it is enough to include European-style derivatives built on underlying risky assets already on the market. In this case, the number of added derivatives coincides with the number of market modes. The problem of replenishing the market has a nonunique solution and the article compares the proposed replenishment method with the existing one.

Keywords: Markov jump process; financial portfolio; self-financing property; market completeness

LOWER BOUND TO ESTIMATION DISTORTION OF A RANDOM PARAMETER FOR A GIVEN AMOUNT OF INFORMATION

• M. M. Lange  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
• A. M. Lange  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: Given probability distribution density with an unknown value of a random parameter, a minimum of the average square distortion for the parameter estimates via the samples of random values as a function of the average mutual information between the samples and the estimates is investigated. This function is produced by inverting a modified rate distortion function as the dependency of the minimal values of the average mutual information on the appropriate values of the average distortion. The obtained smallest average square distortion as the function of the average mutual information is independent on an estimation form and this function yields the lower bound to the average distortion for the fixed values of the amount of information. The above relation is the bifactor fidelity decision criterion that allows one to compare various estimation functions by their efficiency in terms of the average distortion redundancy relative to the lower bound when the entropy of the quantized estimates is fixed.

Keywords: probability distribution density; data sample; parameter estimate; square distortion; mutual information; rate distortion function; lower bound; redundancy

STOCHASTIC PATH LOSS MODEL IN 5G NETWORK DEPLOYMENT SCENARIOS: A STUDY BASED ON 3GPP TR 38.901

• E. D. Makeeva  RUDN University, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation, V. A. Trapeznikov Institute of Control Science of the Russian Academy of Sciences, 65 Profsoyuznaya Str., Moscow 117997, Russian Federation
• I. A. Kochetkova  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 fifth-generation (5G) and beyond networks will utilize radio frequencies in the terahertz spectrum, enabling extremely high data transmission rates. However, signal loss may occur when signals pass through obstacles, making it crucial to simulate signal propagation using stochastic geometry and apply up-to-date models for signal attenuation. The 3GPP TR 38.901 specification includes models that describe signal attenuation in various 5G network scenarios using empirical formulas. Nevertheless, simpler formulas are typically employed to create models based on stochastic geometry. The authors present the cumulative distribution function for path loss at random user locations according to the scenarios described in 3GPP TR 38.901. In numerical examples, it is shown that the difference in values with the simplified formula can be significant and lead to underestimation of the network's capacity

Keywords: wireless network; 5G; 3GPP TR 38.901; path loss; line-of-sight (LOS); non-line-of-sight (NLOS); stochastic geometry

ASSESSING THE CHARACTERISTICS OF 5G/6G "NEW RADIO" SYSTEMS WITH USER'S MACRO- AND MICROMOBILITY

• D. Yu. Ostrikova  RUDN University, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation
• E. S. Golos  RUDN University, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation
• V. A. Beschastnyi  RUDN University, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation
• E. A. Machnev  RUDN University, 6 Miklukho-Maklaya Str., Moscow 117198, Russian Federation
• V. S. Shorgin  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
• Yu. V. Gaidamaka  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

Abstract: The performance of 5G/6G cellular "new radio" systems is typically evaluated using static user location and perfectly directional antennas which are justified in the case of regular synchronisation between the user equipment (UE) and the base station (BS). However, in the case of high energy-efficient UEs with limited RedCap functionality, BS is less likely to get information about the quality of the signal received by the device which changes when the UE moves. This leads to the need to investigate the dynamics of the performance indicators of systems with RedCap UEs over time. In the paper, tools of stochastic geometry and random walk theory are used to analyze the spectral efficiency depending on the distance between the BS and the UE and the directionality of the UE antenna at a random moment of time. A numerical experiment has shown that macromobility has a significant impact on the spectral efficiency, the impact of micromobility is smaller and appears only at short time intervals, while the size of the phased antenna array on the BS side practically does not affect the obtained result.

Keywords: 5G new radio; mmWave; sub-THz; micromobility; macromobility; spectral efficiency

ON SINGLE-THRESHOLD QUEUE MANAGEMENT IN A QUEUING SYSTEM WITH IMPATIENT CUSTOMERS

• 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 results of a theoretical study of a managed queuing system of M/M/k type with impatient customers and single-threshold queue management are presented. The task of optimizing single-threshold queue management is set, the essence of which is to calculate for the queue length a certain threshold value that maximizes a given objective function. In the system under study, a customer leaves the system unattended if the waiting time in the queue (or the service time on the device) exceeds a certain time interval of random length distributed according to an exponential law with a given parameter. A cost function is used as an indicator of the effectiveness of queue management (objective function) which takes into account the losses per unit of time due to system technical maintenance, rejection of customers at the entrance of the system, and leaving of customers until the end of the service. A method for solving the problem of maximizing the cost objective function on a set of single-threshold queue controls and an algorithm for guaranteed calculation of the optimal threshold are proposed.

Keywords:  queuing system; impatient customers; queue management

ON THE GENERATION OF SYNTHETIC FEATURES BASED ON SUPPORT CHAINS AND ARBITRARY METRICS WITHIN THE FRAMEWORK OF A TOPOLOGICAL APPROACH TO DATA ANALYSIS. PART 2. EXPERIMENTAL TESTING ON PHARMACOINFORMATICS PROBLEMS

• I. Yu. Torshin  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: Consideration of precedent relationships between features and a target variable in the form of sets of Boolean lattice elements indicates the possibility of generating synthetic features using metric distance functions. Approaches to (i) assessing the relevance ("informativeness") of metrics in relation to the problems being solved; (ii) generating; and (iii) selecting synthetic features that are more informative than the original feature descriptions are formulated. The results of topological analysis of 2400 samples of "molecule-property" data from ProteomicsDB made it possible to obtain fairly effective algorithms for predicting the properties of molecules (rank correlation in cross-validation is 0.90 ± 0.23). Using this sample of problems, metrics have been established that most often generate informative synthetic features: maximum Kolmogorov deviation, "oblique" distance, and Lp, Renyi, and von Mises metrics. To solve the studied set of problems, the advantage of polynomial correctors compared to neural network and random forest correctors is shown.

Keywords: topological data analysis; lattice theory; algebraic approach of Yu. I. Zhuravlev; pharmacoinformatics

IDENTIFICATION OF CAUSE-AND-EFFECT RELATIONSHIPS WHEN COVERING CAUSES

• 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

Abstract: The tasks of identification of cause-and-effect relationships are of great importance in medical diagnostics, finding the root causes of failures in software and hardware systems, and information security. The explainability of the formed conclusions obtained as a result of complex calculations using artificial intelligence methods is most often realized using causal relationships. The paper investigated the possibility of identification of cause-and-effect relationships in cases where the cause is in an inseparable object available for observation. In such cases, it is said that the "cause" property is covered by an object in which other data properties are present. Effects of causes appear in other information spaces. The cause-and-effect identification problem is investigated in the presence of other random data not related to the relationship generated by the cause-and-effect relationship. The model of deterministic cause-and-effect relationship is considered in the presence of a significant number of randomly occurring properties that are not related to the causal effect of some properties on others.

Keywords: artificial intelligence; computer data analysis; cause and effect; covering causes

IMAGE DECOMPOSITION WITH DISCRETE WAVELET TRANSFORM TO DESIGN A DENOISING NEURAL NETWORK

• A. S. Kovalenko  Institute of Mathematics, Mechanics, and Computer Science named after 1.1. Vorovich, Southern Federal University, 105/42 Bolshaya Sadovaya Str., Rostov-on-Don 344006, Russian Federation

Abstract: Reducing noise in digital images is one of the most common tasks in image processing. At the moment, noise reduction approaches based on the applying of convolutional neural networks are widely used. In this case, as a rule, model training is based on minimizing the error function between the result of the network operation and the expected reference image and, additionally, various representations of the two-dimensional image signal and their properties are not used to optimize the training of noise reduction network architectures. The paper proposes an approach to training neural networks to suppress noise. The described approach is based on the usage of the N-fold fast Haar wavelet transform. This representation of a discrete image signal allows one to discard the classical architecture of the autoencoder and to use only its part that encodes the signal which leads to a significant reduction in model parameters and speeds up the network.

Keywords: neural networks; deep learning; image denoising; image processing

ON THE APPLICATION OF GENERATIVE MODELS IN THE E-LEARNING SYSTEM OF MATHEMATICAL DISCIPLINES

• A. V. Bosov  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation
• A. V. Ivanov  Federal Research Center "Computer Science and Control" of the Russian Academy of Sciences, 44-2 Vavilov Str., Moscow 119333, Russian Federation

Abstract: The existing tools for individual learning trajectory dynamic design are complemented by the generating technology of certification tasks and exam tickets. A set of exam tickets specially prepared by experts in the university course of the theory of functions of a complex variable was used as a source of high-quality, balanced sets of tasks. This significant training array of high-quality attestation tasks has significantly expanded the available data created at previous stages. The purpose ofthe performed research was to create methods that allow taking into account the experts' knowledge embedded in the available set oftasks. The implemented generation model when processing educational content uses as parameters the attributes assigned by experts to tasks: topic, complexity, and formed competencies. Two generation methods are proposed. The first one, probabilistic, uses only the frequency characteristics of the training set, approximating the probability distribution. The second one is based on generative-adversarial neural networks. Particular attention is paid to the discussion of the difficulties of the network implementation, including those related to the specific nature ofthe generative model.

Keywords: e-learning system; educational content; machine learning; generative models; computer simulation; generative-adversarial networks

OBJECT TRANSFORMATIONS OF THE FIRST AND SECOND ORDER IN A LEXICOGRAPHIC INFORMATION SYSTEM

• 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: The theoretical foundations of the design of information technologies used for the integration of bilingual dictionaries and parallel corpora are considered. The description of the first outcomes of the creation of the third level of object transformations classification in the subject domain of informatics, which is supposed to be used in creating the lexicographic information system providing integration, is given. All the entities of informatics are divided into two global classes: objects and their transformations. For each such class, its own classification is constructed. Previously, the two upper levels of the object transformation classification in the subject domain have been described. The present paper discusses the third level of this classification. The basis for the construction of its highest level was the division of the subject domain of informatics into media (mental, sensory, digital, and a number of other media), each of which by definition includes objects of the same nature. The Solomonick's typology of sign systems served as the basis for constructing the second level of the object transformation classification. The aim of the paper is to systematize object transformations of the first and second orders at the third level of this classification. The basis for systematization is the medium version of the Ackoff's hierarchy.

Keywords: subject domain objects; object transformations; classification; data; information; knowledge; lexicographic information system