Vasily Osipov

Vasily Osipov
St. Petersburg institute for informatics and Automation RAS

Ph.D., Dc.Sci., Prof.

About

49
Publications
8,635
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
129
Citations

Publications

Publications (49)
Article
Full-text available
This article addresses an approach to intelligent safety control of passengers on escalators. The aim is to improve the accuracy of detecting threatening situations on escalators in the subway to make decisions to prevent threats and eliminate the consequences. The novelty of the approach lies in the complex processing of information from three typ...
Article
Full-text available
The paper presents the results of statistical data from open sources on the development of the COVID-19 epidemic processing and a study сarried out to determine the place and time of its beginning in Russia. An overview of the existing models of the processes of the epidemic development and methods for solving direct and inverse problems of its an...
Article
Full-text available
Introduction: When substantiating promising architectures of streaming recurrent neural networks, it becomes necessary to assess their stability in processing various input signals. For this, stability diagrams are constructed containing the results of simulation for each of the nodes of these diagrams. Such an estimation can be time-consuming and...
Article
The paper considers a problem of forecasting of news feeds content. Analysis of existing approaches to this problem solution reveals the need for development of methods with enhanced forecasting capabilities. A method is proposed with expanded accounting for space and time relationships of the processed data. The method is revealed through an examp...
Article
Full-text available
Ambient Intelligence System (AmIS) can be constructed using data collected from Internet of Things (IoT). In this paper, the IoT data collection problem is studied for AmIS with dynamic structure and dynamic behavior of participants (devices), where constraints on resources consumption and performance are essential. A novel technology is proposed,...
Article
Introduction: High hopes for a significant expansion of human capabilities in various fields of activity are pinned on the creation and use of highly intelligent robots. To achieve this level of robot intelligence, it is necessary to successfully solve the problems of predicting the external environment and the state of the robots themselves. Solut...
Article
Full-text available
The problem of neural network forecasting of processes with changing laws of their behavior and imperfection of time-series samples is considered on the example of analysis of urban traffic flows. The goal is to improve the accuracy of such forecasting. To achieve this goal, we analyze the applicability of self-learning recurrent neural networks wi...
Article
Full-text available
This article addresses the monitoring problem of the telecommunication networks. We consider these networks as multilevel dynamic objects. It shows that reconfigurable systems are necessary for their monitoring process in real life. We implement the reconfiguration abilities of the systems through the synthesis of monitoring programs and their exec...
Chapter
In the chapter, the problem of cyber-physical systems monitoring based on the processing and evaluation of the data received from the observed objects is considered. The main attention is paid to finding a suitable way to reduce the synthesis complexity of the observed objects models and monitoring processes. The authors propose a multilevel automa...
Article
Full-text available
Practically all human activities depend on technical systems, which consists of a multitude of dynamically distributed objects. In order to control these systems, it is necessary to build and periodically rebuild models of objects, which consist of elements and connections between them and describes the object's state in time and space. Due to a la...
Chapter
In the paper the problem of high computational complexity of synthesis is discussed. Existing models and methods of synthesis don’t allow build models of biological objects and systems. The complexity can be significantly reduced due to considering multilevel objects models instead of single level models. The new problem statement for multilevel sy...
Chapter
The features of recurrent neural networks with controlled elements are considered. The functions of these networks for controlled associative processing of signals are refined. A number of the space-time structures of such networks are analyzed. Among them, the neural networks with one-, two-, and three-level structures of layers are investigated....
Article
The concept of cognitive monitoring is defined. Some possible approaches to the construction of cognitive monitoring systems are considered and their generalized structure is described. The concept of a cognitive monitoring machine is introduced. A cognitive architecture approach to design monitoring systems that features the generation of on-deman...
Chapter
Traffic congestions have a strong impact on the life of modern cities. Reducing congestion is one of the main concerns of private, urban and public institutions. Much effort has been devoted to the scientific research of this problem. One of the areas of these studies is the prediction of congestion. The forecast helps to distribute traffic on urba...
Chapter
Streaming recurrent neural networks with linear and spiral space-time structures are analyzed. Formal aspects of the construction and operation of such networks are considered. Four types of synapses are identified in such networks. One of them is the track synapses, which ensure the advancement of signals over the network. Three other types are sy...
Article
The possibilities of signal binding in recurrent neural networks with controlled elements are investigated. It is shown that a variety of dynamic space–time structures with new associative properties can be formed in the framework of such networks. A comparative analysis of the properties of linear, spiral single-level and multilevel structures of...
Conference Paper
Nowadays big volumes of medical data are accumulated. So the problem of analysis of these data and mining linked logical structures, defining internal data semantics is an actual one. Solution of this problem allows solve the problem of optimizing intelligent context search. In the article an approach for solving this problem for analyzing processe...
Conference Paper
In the article an approach to solving the problem of inductive synthesis of the models of biological systems according to clinical trials is suggested. Suggested approach to inductive synthesis of biological models on the base of results of clinical trials allows essentially decrease computational complexity of this problem. Formalization of biolog...
Article
Full-text available
This paper provides an analysis of the present state in the field of protection against false information in computer networks and formulates current problems related to this protection. An approach to assessing protection activities on the basis of the Markov chain of the disinformation process is proposed. The architecture of a future system of d...
Article
Full-text available
The structure of a cognitive machine based on a recurrent neural network with controlled elements is considered. The basic functions of this machine are defined and analyzed. The expanded capabilities of the cognitive machine to solve creative tasks are discussed. The results of modelling are given.
Conference Paper
A model of recurrent neural network with controlled synapses is considered. Issues of improvement of space-time structures of neural networks are discussed. Possible impacts on associative interactions between signals in recurrent neural networks are studied by means changing their spatial parameters depending on the current states of network layer...
Article
The problem of automatic synthesis of action programs for intelligent robots in a permanently changing environment is considered. Capabilities of the deductive synthesis of cyclic and self-replicating programs are investigated. New algorithms for synthesis of such action programs for a wide range of possible conditions are suggested. Results of thi...
Article
Full-text available
Purpose: Increasing the functionality of associative intelligent machines for dynamic signal processing. Methods: In order to give new features to associative intelligent machines, these machines are constructed with two signaling systems based on recurrent neural networks with operated synapses. Special solutions changed the spatial characteristic...
Article
Full-text available
The capabilities of recurrent neural networks with operated synapses with deletion of outdated information are investigated. Problem of estimating the storage capacity of these networks is solved. It is noted that for its decision because of specifics of rules of storing of information known methods are not suitable. They are not focused on an asse...
Article
Full-text available
Purpose. Search approaches to eliminate excess storage and to empower recognition of dynamic signals in recurrent neural networks. Methods: The proposed approach is based on well-known models and methods of information processing in recurrent neural networks (RNN) with operated synapses. To justify the proposed approach used method of mathematical...
Article
The advanced method of information processing in associative intelligent machines (AIM) constructed on the basis of the recurrent neural networks (RNN) with operated synapses is considered. It is offered to allocate AIM with three signals systems. The model of operated synapses is specified. Dependences between forms and the directions of associati...
Article
Full-text available
Approach to expand the functionality of associative intelligent systems is considered. Control features an associative interaction of signals in these systems and giving them a second signal system are disclosed. Proposals for the analog implementation of such systems are formulated
Article
Full-text available
Technique for optimization of visual environment for spatial modeling is considered. It was implemented. The use of the technique in intelligent subsystem of visual environment for spatial modeling is proposed.
Article
Full-text available
Purpose: To find flexible approaches providing prompt identification of an expedient period of revision of information security measures depending on a current situation. Methods: The mathematical apparatus of Markov processes has been used to estimate possibilities of finding information security actions. Searching this period a process structure,...
Article
Full-text available
There has been proposed an approach to provide operated synapses recurrent neural network with the second signal system allowing to process signals regardless of direct perception of reality. A model of such a network with expanded functionality has been considered. There have been presented simulation results showing a possibility of essential inc...
Article
Full-text available
There has been presented an approach to substantiate information security measures with account of value of protected information resources. There has been consid- ered a model of value of these resources. There has been presented mathematical formulation and an algorithm for solving the problem of searching for expedient infor- mation security mea...
Article
Full-text available
One of the important scientific and technological problems of our time is to establish a "thinking" systems. Among the well-known approaches to solving this problem deserves special attention the development of methods and systems for intellectual associative neural information processing. However, while these methods and systems are not perfect, a...
Article
Full-text available
The method of adjustment of associative intelligent system on entrance signals taking into account current loading of the system, expanding possibilities on processing of the information is considered. The mathematical formulation and algorithm of the decision of a problem is resulted. Results of modelling are reflected.
Article
Full-text available
Capabilities of a recurrent neural network with spiral structure of layers are investigated. The algorithm of its functioning is considered. The results of modeling are given. It is shown that the network demonstrates selectivity of information processing at counter advancement of sets of individual images along the layers. Due to this advancement...
Article
Full-text available
An approach to erase outdated information in associative intelligent systems (AIS) based on partial reflection signals from the layers of neural networks is offered. A mathematical model of a neural network with such an erase is made. Results of modeling are presented. It is shown that the proposed approach improves the stability of the AIS.
Article
Full-text available
The approach to the sustainability of recurrent neural networks (RNN) with controlled synapses was researched. Model synapses and stability of RNN were clarified during this research. Possible options for inhibition of neurons in the RNN were investigated. The results of mathematical modeling are presented. Practical recommendations on the sustaina...
Article
Full-text available
An approach to optimization of associative Intelligent Systems (AIS). Introduced indicator of their associative performance. According to this indicator investigated the benefits of a number of logical structures within the AIS bilayer recurrent neural networks with managed synapses. The results of mathematical modeling. Practical recommendations f...
Article
Full-text available
The constructional to approach of the perspective intelligent machine on the basis of a special recurrent neural network is contemplate. Operation principles of this machine and its architecture is disclosed. The state transition analysis of the intelligent machine at processing the input compound signals is presented.
Conference Paper
This paper considers the problem of opposing the terroristic threat. It covers one of the aspects, which it was investigated within a complex project that examined the development of the terroristic threat monitoring system based on geoinformation technologies. A biographical data analysis has been made for 93 known terrorists. The analysis has res...
Article
Full-text available
1 ABSTRACT This paper introduces the problems of substantiation and development of the new generation of Vessel Traffic Control Systems (VTCS), together with the potential of intelligent-based Geographic Information System (GIS). Nowadays, it is almost impossible to imagine not only World economics life but also every local economics and social lif...
Article
An approach is considered to develop a system of models for the optimal control of radiotransmitting devices (RTD). The system is oriented for the wide spectrum of possible practical conditions under which automated radiostations (RS) are functioning. A system of problems for the optimal control of RTDs is formulated based on the selected versions...
Article
The developed model permitted to distribute radioelectronic means (REM) to perform user's requirements with account of structural peculiarities of control system (CS) of REM, functioning rules of it's elements, temporal characteristic changes. Structure of REM C5 was presented as a graph, the peaks of which turned out to be dishwasher, operators, R...
Article
One of the ways to improve decametric communications (DCMC) is to increase the degree of its automation by designing a computer-aided DCMC control system (DCMC CS). Successful solution of this task depends both on the available computer power and the use of special problem-solution methods inherent in DCMC CS. Experience of controlling DCMC testifi...

Questions

Question (1)
Question
If you originally shared by all neurons at excitatory and inhibitory, it significantly limits the ability of a neural network with a given structure. Why not use a one type of neurons. Are there any known electrical models of neurons, providing both intensification and weakening of synaptic connections?

Network

Cited By

Projects

Project (1)
Project
It is an extension of the functionality of these machines.