×

You are using an outdated browser Internet Explorer. It does not support some functions of the site.

Recommend that you install one of the following browsers: Firefox, Opera or Chrome.

Contacts:

+7 961 270-60-01
ivdon3@bk.ru

  • Application of language neural network models for malware detection

    The growing popularity of large language models in various fields of scientific and industrial activity leads to the emergence of solutions using these technologies for completely different tasks. This article suggests using the BERT, GPT, and GPT-2 language models to detect malicious code. The neural network model, previously trained on natural texts, is further trained on a preprocessed dataset containing program files with malicious and harmless code. The preprocessing of the dataset consists in the fact that program files in the form of machine instructions are translated into a textual description in a formalized language. The model trained in this way is used for the task of classifying software based on the indication of the content of malicious code in it. The article provides information about the conducted experiment on the use of the proposed model. The quality of this approach is evaluated in comparison with existing antivirus technologies. Ways to improve the characteristics of the model are also suggested.

    Keywords: antivirus, neural network, language models, malicious code, machine learning, model training, fine tuning, BERT, GPT, GPT-2

  • Using the capabilities of GPUs for mathematical calculations

    The present paper examines the actual problem of using graphics processing units (GPUs) in computing processes that are traditionally performed on central processing units (CPUs). With the development of technology and the advent of specialized architectures and libraries, GPUs have become indispensable in areas requiring intensive computing. The article examines in detail the advantages of using GPUs compared to traditional CPUs, justifying this with their ability to process in parallel and high throughput, which makes them an ideal tool for working with large amounts of data.are accidents caused by violations of rules and regulations at work sites, among them cases related to non-compliance with the rules of wearing protective helmets. The article examines methods and algorithms for recognizing protective helmets and helmets, and assesses their effectiveness.

    Keywords: graphics processors, GPU, CUDA, OpenCL, cuBLAS, CL Blast, rocBLAS, parallel data processing, mathematical calculations, code optimization, memory management, machine learning, scientific research

  • Optimization of the dense matrix multiplication procedure for shared memory systems

    The study presents an extensive analysis of methods for low-level optimization of the matrix multiplication algorithm for computing systems with shared memory. Based on a comparison of various approaches, including block optimization, parallel execution with OpenMP, vectorization with AVX and the use of the Intel MKL library, significant improvements in the performance of the resulting software implementations are revealed. In particular, block optimization reduces the number of cache misses, parallelism effectively uses multicore, and vectorization and Intel MKL demonstrate maximum acceleration due to more efficient software optimizations. The obtained results emphasize the importance of careful selection of optimization methods and their compliance with the architecture of the computing system in order to achieve the required performance parameters of the designed software.

    Keywords: low-level optimization, block optimization, parallel execution, OpenMP, vectorization, AVX, Intel MKL, performance, benchmarking, matrix multiplication

  • The use of neural network detectors to prevent accidents at work

    This article discusses the features of the use of neural network detectors in the tasks of recognizing protective helmets and helmets. Nowadays, workplace safety is becoming an increasingly relevant topic, especially in industries with a high level of injuries. There are accidents caused by violations of rules and regulations at work sites, among them cases related to non-compliance with the rules of wearing protective helmets. The article examines methods and algorithms for recognizing protective helmets and helmets, and assesses their effectiveness.

    Keywords: convolutional neural network, object recognition, protective helmets. helmets, SSD, YOLOv5, Faster R-CNN, machine learning, deep learning, image classification

  • Algorithms for the movement of a mobile robot with the construction of a real-time terrain map

    This article describes the implementation of an orientation software package for a mobile robot with the construction of a terrain map and its subsequent analysis. As part of the work, a software module was developed for a mobile robot using a laser rangefinder (lidar), a Raspberry Pi 3B+ microcomputer with a ROS robotic operating system installed on it is used to obtain data from a laser rangefinder. The algorithm of movement of a mobile robotic complex in space with the construction of a terrain map in real time is described. Such complexes are currently widely used, they can significantly reduce the need for human participation in heavy and dangerous work.

    Keywords: mobile robot, robotic complex, laser rangefinder, lidar, ultrasonic sensor, vision system

  • Using Euler angles in inertial navigation systems

    This article discusses the features of using Euler angles in the process of determining the position of an object in space. The material presented here will help to study in detail the use of Euler angles when working with inertial navigation sensors and facilitate the understanding of the construction of systems using such devices. Today, such embedded automated systems are used in a wide variety of fields: in aircraft modeling, in medicine, military equipment, and agriculture, and the scale of their use is growing every year. Devices that track the position of the body in space are widely used in such systems as drones, mobile robots, remote-controlled manipulators. These devices can significantly reduce the need for human participation in heavy and dangerous work.

    Keywords: Euler angles, accelerometer, gyroscope, magnetometer, object orientation, inertial navigation sensors, yaw, pitch, roll

  • Methods and tools for load balancing in heterogeneous computing systems

    Currently, there is a rapid development of computing systems, including systems that include various types of devices. Such systems are called heterogeneous or heterogeneous. In addition to traditional processors, such a system may include graphics processing units, programmable gate arrays, and other devices. One of the most pronounced problems of such systems is the complexity of distributing the computational load between the nodes of the computing system. This paper describes the types of balancing and methods for distributing computational load in heterogeneous computing systems. An example of load balancing in a real heterogeneous computing system is described.

    Keywords: heterogeneous system, heterogeneous system, load balancing, load distribution, computing cluster