Search
Filters
total: 1150
displaying 1000 best results Help
Search results for: restricted boltzmann machine
-
Video of LEGO bricks on conveyor belt - Special Brics
Open Research DataThe set contains videos of LEGO bricks (special bricks, with additional connectors etc.) moving on a white conveyor belt. The videos were prepared for training neural network for recognition of LEGO bricks. The bricks were separated as much as possible and in most cases they should not overlap. The images were taken from different sides by stationary...
-
Video of LEGO bricks on conveyor belt - Wide Brics
Open Research DataThe set contains videos of LEGO bricks (wide bricks, with each side having more than 1 stud) moving on a white conveyor belt. The videos were prepared for training neural network for recognition of LEGO bricks. The bricks were separated as much as possible and in most cases they should not overlap. The images were taken from different sides by stationary...
-
Video of LEGO bricks on conveyor belt - minifigures, animals, plants and accessories
Open Research DataThe set contains videos of LEGO bricks (minifigures, animals, plants and accessories) moving on a white conveyor belt. The images were prepared for training neural network for recognition of LEGO bricks. The bricks were separated as much as possible and in most cases they should not overlap. The images were taken from different sides by stationary camera...
-
Video of LEGO bricks on conveyor belt - Narrow Brics
Open Research DataThe set contains videos of LEGO bricks (narrow bricks, with on side no wider than 1 stud) moving on a white conveyor belt. The videos were prepared for training neural network for recognition of LEGO bricks. The bricks were separated as much as possible and in most cases they should not overlap. The images were taken from different sides by stationary...
-
Vertical vibration reduction and audible sound analysis in surface grinding with electroplated tools
PublicationOne of the first approaches to the development of a grinding process monitoring system based on audible sound sensors is presented in the paper. Electroplated diamond tools (abrasive D64 and D107) were used in a modified single-disc lapping machine configuration for flat grinding of ceramics (Al2O3). The main aim of the machine modification was to reduce the vertical vibration in order to decrease the tool wear and to increase...
-
Asking Data in a Controlled Way with Ask Data Anything NQL
PublicationWhile to collect data, it is necessary to store it, to understand its structure it is necessary to do data-mining. Business Intelligence (BI) enables us to make intelligent, data-driven decisions by the mean of a set of tools that allows the creation of a potentially unlimited number of machine-generated, data-driven reports, which are calculated by a machine as a response to queries specified by humans. Natural Query Languages...
-
Adam Dąbrowski dr inż.
PeopleAdam Dabrowski has obtained a PhD in mechanical engineering from Gdańsk University of Technology and MSc. degree in mechatronics from Technische Universität Hamburg. He has an industry experience in Institute of Aviation Engineering Design Center (Warsaw, Poland) and SICK AG (Hamburg, Germany). Additionally, as an assistant at Gdansk University of Technology he teaught courses on mechanics, space mechanisms and dynamics of space...
-
Deep Learning Basics 2023/24
e-Learning CoursesA course about the basics of deep learning intended for students of Computer Science. It includes an introduction to supervised machine learning, the architecture of basic artificial neural networks and their training algorithms, as well as more advanced architectures (convolutional networks, recurrent networks, transformers) and regularization and optimization techniques.
-
Uczenie maszynowe do samoorganizacji systemów rozproszonych w zastosowaniach gospodarczych
PublicationW pracy omówiono uczenie maszynowe do samoorganizacji systemów rozproszonych w zastosowaniach gospodarczych ze szczególnym uwzględnieniem sieci neuronowych do predykcji finansowych oraz szacowania ratingu przedsiębiorstw. Oprócz sieci neuronowych, istotną rolę w przygotowaniu i testowaniu informatycznych systemów finansowych może pełnić programowanie genetyczne. Z tego powodu omówiono uczenie maszynowe w aplikacjach konstruowanych...
-
Obserwatory prędkości dla bezczujnikowego sterowania maszynami prądu przemiennego
PublicationPrzedstawiono model matematyczny uogólnionej maszyny elektrycznej rozszerzony przez wprowadzenie dodatkowych zmiennych. Na podstawie modelu rozszerzonego opracowano strukturę obserwatora prędkości maszyny uogólnionej. Zaprezentowano struktury obserwatorów prędkości dla poszczególnych rodzajów maszyn prądu przemiennego. Pokazano, że prędkość kątową wirnika można odtwarzać dla różnych typów maszyn stosując odpowiednie zmienne stanu...
-
A Measurement-Based Approach for Speed Control of Induction Machines
PublicationThis paper presents an approach to design a measurement-based controller for induction machines. The proposed control approach is motivated by the fact that developing an appropriate mechanical model of such induction machines is a challenging task. Since our proposed control methodology is only on the basis of measured data, the controller design does not require any information about the model of the mechanical part. The control...
-
Frequency Response Functions for Sandvik CoroBore 825 XL boring tool, 733 mm
Open Research DataFrequency Response Functions (FRF) obtained for Sandvik CoroBore® 825 XL mounted at WHN 13-15 TOS VARNSDORF CNC machine at the minimum extension of the spindle adaptor. Tool set to bore a hole with a diameter 733mm
-
Olgun Aydin Dr
PeopleOlgun Aydin finished his PhD by publishing a thesis about Deep Neural Networks. He works as a Senior Data Scientist in PwC Poland, gives lectures in Gdansk University of Technology in Poland and member of WhyR? Foundation. Olgun is a very big fan of R and author of the book called “R Web Scraping Quick Start Guide” , two video courses are called “Deep Dive into Statistical Modelling using R” and “Applied Machine Learning and Deep...
-
Roller burnishing
PublicationThe paper shows the process of roller burnishing as a method of finishing machine components. Based on own research the author presents the effects of the roller burnishing process to increase the hardness and residual stress as well as the wear and fatigue strength of burnishing components.
-
Human Feedback and Knowledge Discovery: Towards Cognitive Systems Optimization
PublicationCurrent computer vision systems, especially those using machine learning techniques are data-hungry and frequently only perform well when dealing with patterns they have seen before. As an alternative, cognitive systems have become a focus of attention for applications that involve complex visual scenes, and in which conditions may vary. In theory, cognitive applications uses current machine learning algorithms, such as deep learning,...
-
Category-Based Workload Modeling for Hardware Load Prediction in Heterogeneous IaaS Cloud
PublicationThe paper presents a method of hardware load prediction using workload models based on application categories and high-level characteristics. Application of the method to the problem of optimization of virtual machine scheduling in a heterogeneous Infrastructure as a Service (IaaS) computing cloud is described.
-
Patryk Ziółkowski dr inż.
PeoplePatryk Ziolkowski is a graduate of the Faculty of Civil and Environmental Engineering at the Gdansk University of Technology, specializing in Building and Engineering Structures. He works as an Assistant Professor at the Department of Engineering Structures. He participated in international projects, including projects for the Ministry of Transportation of the State of Alabama (2015), he is also the winner of a grant from the Kosciuszko...
-
Koncepcja systemu wspomagania decyzji nawigatora statku opartego na ewolucyjnym planowaniu manewrów antykolizyjnych
PublicationArtykuł przedstawia koncepcję systemu wspomagania decyzji nawigatora statku opartego na wątkach badań prowadzonych wcześniej przez autora. System będzie rozszerzał funkcjonalność systemów dotychczasowych o możliwość szczegółowego planowania bezpiecznej trajektorii statku na wodach zamkniętych, z dużą liczbą statków obcych i ograniczeniami toru wodnego. Artykuł zawiera dyskusję możliwych podejść do planowania manewrów, optymalizacji...
-
From Linear Classifier to Convolutional Neural Network for Hand Pose Recognition
PublicationRecently gathered image datasets and the new capabilities of high-performance computing systems have allowed developing new artificial neural network models and training algorithms. Using the new machine learning models, computer vision tasks can be accomplished based on the raw values of image pixels instead of specific features. The principle of operation of deep neural networks resembles more and more what we believe to be happening...
-
Lessons learned from developing an Industry 4.0 mobile process management system supported by Artificial Intelligence
PublicationResearch, development and innovation (RDI) projects are undertaken in order to improve existing, or develop new, more efficient products and services. Moreover, the goal of innovation is to produce new knowledge through research, and disseminating it through education and training. In this line of thinking, this paper reports and discusses the lessons learned from the undertaken project, regarding three areas: machine learning...
-
Identification and non-integer order modelling of synchronous machines operating as generator
PublicationThis paper presents an original mathematical model of a synchronous generator using derivatives of fractional order. In contrast to classical models composed of a large number of R-L ladders, it comprises half-order impedances, which enable the accurate description of the electromagnetic induction phenomena in a wide frequency range, while minimizing the order and number of model parameters. The proposed model takes into account...
-
Investigation of the Effect of The Temperature and Magnetization Pattern on Flux Density, Instantaneous Torque, Unbalanced Magnetic Forces of a Surface Inset PMM
PublicationElectrical machines utilized in domestic applications such as ceiling fans should have low losses and cost. Permanent magnets are used instead of rotor excitation to reduce losses. Therefore, not only the losses of the rotor winding are eliminated, but also the efficiency of the machine is increased. A surface inset consequent pole (SICP) machine has also been used to reduce costs. Because less magnets are utilized in this structure....
-
A subdomain model for armature reaction field and open‐circuit field prediction in consequent pole permanent magnet machines
PublicationIn this paper, the machine quantity, such as electromagnetic torque, self and mutual inductances, and electromotive force, is analytically calculated for non-overlapping winding consequent pole slotted machine for open-circuit field and armature reaction. The sub-domain approach of (2-D) analytical model is developed using Maxwell's equations and divide the problem into slots, slot-openings, airgap and magnets region, the magnet...
-
Valence and lowest Rydberg electronic states of phenol investigated by synchrotron radiation and theoretical methods
PublicationWe present the experimental high-resolution vacuum ultraviolet (VUV) photoabsorption spectra of phenol covering for the first time the full 4.3–10.8 eV energy-range, with absolute cross sections determined. Theoretical calculations on the vertical excitation energies and oscillator strengths were performed using time-dependent density functional theory and the equation-of-motion coupled cluster method restricted to single and double...
-
ELECTRON ATTACHMENT TO HETEROGENEOUS MOLECULAR CLUSTERS
PublicationHeterogeneous clusters are unique model species enabling studies of bulk and surface processes by vacuum restricted methods. This is of particular importance for the reactions of low-energy electrons which have extremely short penetration depth in bulk environments. This contribution overviews our recent experiments with low-energy electrons and clusters prepared by different techniques. First, the deposition of molecules on argon...
-
Minimal parameter implicit solvent model for ab initioelectronic-structure calculations
PublicationAbstract - We present an implicit solvent model for ab initio electronic-structure calculations which is fully self-consistent and is based on direct solution of the nonhomogeneous Poisson equation. The solute cavity is naturally defined in terms of an isosurface of the electronic density according to the formula of Fattebert and Gygi (J. Comput. Chem., 23 (2002) 662). While this model depends on only two parameters, we demonstrate...
-
Electron-impact ionization of fluoromethanes – Review of experiments and binary-encounter models
PublicationExperiments and recommended data on electron-impact ionization of methane and fluoromethanes (CH3F, CH2F2, CHF3, CF4) are reviewed and compared with binary-encounter models (Gryzinski’s, ´ Deutsch and Märk’s, and Kim and Rudd’s). A good agreement between recent experiments and the two latter classical-like models is shown. Kim and Rudd’s model (calculated presently in the restricted HartreeFock 6-31**G orbital basis) predicts well...
-
Child-Robot Interaction Studies During COVID-19 Pandemic
PublicationThe coronavirus disease (COVID-19) pandemic affected our lives deeply, just like everyone else, the children also suffered from the restrictions due to COVID-19 affecting their education and social interactions with others, being restricted from play areas and schools for a long time. Although social robots provide a promising solution to support children in their education, healthcare, and social interaction with others, the precautions...
-
Trustworthiness and Subversion in Large Service- Oriented Multi-Agent Systems under Virtual Anonymity and Blind Provider Selection
PublicationTo promote honest service provision in multi-agent systems, a Computational Trustworthiness and Rating scheme collects service ratings and computes agents' trustworthiness levels (TLs). Studies of existing schemes fail to reflect closed-loop dynamics of TLs arising from the mutual influence of agents' behavior and TLs, and to capture relevant properties. Recent simulative and analytical models produce results tied to a particular...
-
Density functional theory calculations on entire proteins for free energies of binding: Application to a model polar binding site
PublicationIn drug optimization calculations, the molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) method can be used to compute free energies of binding of ligands to proteins. The method involves the evaluation of the energy of configurations in an implicit solvent model. One source of errors is the force field used, which can potentially lead to large errors due to the restrictions in accuracy imposed by its empirical nature....
-
Tadeusz Widerski dr inż.
PeopleIn 1998, he passed his secondary school-leaving examination and graduated from the Mechanical Technical School with a machine-building profile in Dobre Miasto. In the years 1998-2003 he studied at the Faculty of Geodesy and Spatial Management of the University of Warmia and Mazury, where he obtained the title of Master of Science in Geodesy and Cartography. From 2005 to 2008 he was a student of the Doctoral Study in geotechnics...
-
Mariusz Deja dr hab. inż.
PeopleAssistant Professor in the Department of Manufacturing Engineering and Automation. He graduated in 1993, obtaining Master degree at the Faculty of Mechanical Engineering, Gdańsk University of Technology in the field of study: Mechanical Engineering, specialization: Technological Process Design. After graduation he worked for the Department of Manufacturing Engineering and Automation and his main research area was associated with...
-
Comparative Analysis of Text Representation Methods Using Classification
PublicationIn our work, we review and empirically evaluate five different raw methods of text representation that allow automatic processing of Wikipedia articles. The main contribution of the article—evaluation of approaches to text representation for machine learning tasks—indicates that the text representation is fundamental for achieving good categorization results. The analysis of the representation methods creates a baseline that cannot...
-
Quantumness in Diagnostics of Marine Internal Combustion Engines and Other Ship Power Plant Machines
PublicationThe article provides proof that the diagnostics of marine internal combustion engines and other ship power plant machines should take into account the randomness and unpredictability of certain events, such as wear, damage, the variations of mechanical and thermal loads, etc., which take place during machine operation. In the article, the energy E, like the other forms (methods) that it can be converted into (heat and work), is...
-
Semantic rules representation in controlled natural language in FluentEditor
PublicationThis paper presents a way of representation of semantic rules (SWRL) in controlled English in order to facilitate understanding the rules by humans interacting with a machine. This approach (implemented in FluentEditor) may be applied in many domains, where the understandability of the rules used to support a decision process is of great importance.
-
A review on analytical models of brushless permanent magnet machines
PublicationThis study provides an in-depth investigation of the use of analytical and numerical methods in analyzing electrical machines. Although numerical models such as the finite-element method (FEM) can handle complex geometries and saturation effects, they have significant computational burdens, are time-consuming, and are inflexible when it comes to changing machine geometries or input values. Analytical models based on magnetic equivalent...
-
Review of the Complexity of Managing Big Data of the Internet of Things
PublicationTere is a growing awareness that the complexity of managing Big Data is one of the main challenges in the developing feld of the Internet of Tings (IoT). Complexity arises from several aspects of the Big Data life cycle, such as gathering data, storing them onto cloud servers, cleaning and integrating the data, a process involving the last advances in ontologies, such as Extensible Markup Language (XML) and Resource Description...
-
Multimedia interface using head movements tracking
PublicationThe presented solution supports innovative ways of manipulating computer multimedia content, such as: static images, videos and music clips and others that can be browsed subsequently. The system requires a standard web camera that captures images of the user face. The core of the system is formed by a head movement analyzing algorithm that finds a user face and tracks head movements in real time. Head movements are tracked with...
-
Analysis of the Capability of Deep Learning Algorithms for EEG-based Brain-Computer Interface Implementation
PublicationMachine learning models have received significant attention for their exceptional performance in classifying electroencephalography (EEG) data. They have proven to be highly effective in extracting intricate patterns and features from the raw signal data, thereby contributing to their success in EEG classification tasks. In this study, we explore the possibilities of utilizing contemporary machine learning algorithms in decoding...
-
Facial feature extraction from a monochromatic picture.
PublicationFace pose determination represents an important area of research in Human Machine Interaction. In this paper, I describe a new method of extracting facial feature locations from a single monochromatic monocular camera for the purpose of estimating and tracking the three dimensional pose of human face and eye-gaze direction.
-
Facial Feature extraction from a monochromatic picture
PublicationFace pose determination represents an important area of research in Human Machine Interaction. In this paper, I describe a new method of extracting facial feature locations from a single monochromatic monocular camera for the purpose of estimating and tracking the three dimensional pose of human face and eye-gaze direction.
-
Steering is an essential feature of non-locality in quantum theory
PublicationA physical theory is called non-local when observers can produce instantaneous effects over distant systems. Non-local theories rely on two fundamental effects: local uncertainty relations and steering of physical states at a distance. In quantum mechanics, the former one dominates the other in a well-known class of non-local games known as XOR games. In particular, optimal quantum strategies for XOR games are completely determined...
-
Testing the Wide-Sense Stationarity of Bandpass Signals for Underwater Acoustic Communications
PublicationThe Underwater Acoustic Communication (UAC) systems work in communication channels characterized by a large variety of multipath propagation conditions that can additionally change over time. Designing a reliable communication system requires knowledge of the transmission parameters of the channel. There is a need for the development of adaptive signaling schemes that would dynamically optimize the signal parameters of the physical...
-
Low-Cost Data-Driven Surrogate Modeling of Antenna Structures by Constrained Sampling
PublicationFull-wave electromagnetic (EM) analysis has become one of the major design tools for contemporary antenna structures. Although reliable, it is computationally expensive which makes automated simulation-driven antenna design (e.g., parametric optimization) difficult. This difficulty can be alleviated by utilization of fast and accurate replacement models (surrogates). Unfortunately, conventional data-driven modeling of antennas...
-
Point cloud based surface analysis of prestressed concrete element in exceeded state of serviceability
PublicationBy definition, serviceability state is characterized by the fact that despite meeting the requirements associated with carrying capacity, the object is not suitable for use. Serviceability limit state refers to factors, such as durability, overall stability, fire resistance, cracking, excessive vibration. Perfect examples of exceeded serviceability state are the improper vibration of a skyscraper, which inhabitants may suffer from...
-
Horizontal velocity field derived from EPN and ASG-EUPOS satellite data on the example of south-western part of Poland
PublicationPresently the determination of the velocity field in the global reference frame is possible by using different space techniques and dense terrestrial networks from global to local and regional scales. However, the reliability of such determinations is strongly limited by the restricted number of unmodeled effects. Some of them are periodic (atmospheric or hydrological effects), some instantaneous (natural or man-made seismicity)...
-
Synthetic strategies in construction of organic macromolecular carrier–drug conjugates
PublicationMany metabolic inhibitors, considered potential antimicrobial or anticancer drug candidates, exhibit verylimited ability to cross the biological membranes of target cells. The restricted cellular penetration ofthose molecules is often due to their highhydrophilicity. One of the possible solutions to this problem is aconjugation of an inhibitor with a molecular organic nanocarrier. The conjugate thus formed should beable to penetrate...
-
Tensile strength test of PE foils modified with zinc oxide and zinc salts
Open Research DataThis dataset contains the results of PE foils mechanical tests, in which the material was evaluated for strength and extensibility at the moment of breaking and based on the same tests preceded by material fatigue in 50 stretching cycles. The principle of the method is to stretch the test sample placed in the clamps of the stretching apparatus at a...
-
Compression test results of fused filamend fabricated polylactyde and polylactyde with carbon fiber filler
Open Research DataThe DataSet contains the results of the mechanical behaviour of cube specimens with dimensions 30 x 30 x 30 cm3. The specimens were additively manufactured with the fused filament fabrication (FFF) method. They used two different materials polylactide (PURE PLA) and polylactide with carbon fibre filler (PLA+CF). The specimens were segregated into three...
-
Artificial Neural Networks as an architectural design tool- generating new detail forms based on the Roman Corinthian order capital
PublicationThe following paper presents the results of the research in the field of the machine learning, investigating the scope of application of the artificial neural networks algorithms as a tool in architectural design. The computational experiment was held using the backward propagation of errors method of training the artificial neural network, which was trained based on the geometry of the details of the Roman Corinthian order capital....