Search
Filters
total: 32965
-
Catalog
displaying 1000 best results Help
Search results for: training , adaptation models , uncertainty , image color analysis , semantics , estimation , predictive models
-
Closer Look at the Uncertainty Estimation in Semantic Segmentation under Distributional Shift
PublicationWhile recent computer vision algorithms achieve impressive performance on many benchmarks, they lack robustness - presented with an image from a different distribution, (e.g. weather or lighting conditions not considered during training), they may produce an erroneous prediction. Therefore, it is desired that such a model will be able to reliably predict its confidence measure. In this work, uncertainty estimation for the task...
-
Wiktoria Wojnicz dr hab. inż.
PeopleDSc in Mechanics (in the field of Biomechanics) - Lodz Univeristy of Technology, 2019 PhD in Mechanics (in the field of Biomechanics) - Lodz Univeristy of Technology, 2009 (with distinction) List of papers (2009 - ) Wojnicz W., Wittbrodt E., Analysis of muscles' behaviour. Part I. The computational model of muscle. Acta of Bioengineering and Biomechanics, Vol. 11, No.4, 2009, p. 15-21 Wojnicz W., Wittbrodt E., Analysis of...
-
Błażej Kochański dr
PeopleBłażej Kochański is an assistant professor at the Department of Statistics and Econometrics at the Faculty of Management and Economics of Gdańsk University of Technology, banking risk expert. He worked for banks in Poland and Europe, as a risk specialist, planning and analysis manager, chief risk officer, supervisory board member and management consultant. He built numerous credit risk management models, successfully managed credit...
-
LDRAW based renders of LEGO bricks moving on a conveyor belt with extracted models
Open Research DataThe set contains renders of LEGO bricks moving on a white conveyor belt. The images were prepared for training neural network for recognition of LEGO bricks. For each brick starting position, alignment and color was selected (simulating the brick falling down on the conveyour belt) and than 10 images was created while the brick was moved across the...
-
LDRAW based renders of LEGO bricks moving on a conveyor belt
Open Research DataThe set contains renders of 5237 LEGO bricks moving on a white conveyor belt. The images were prepared for training neural network for recognition of LEGO bricks. For each brick starting position, alignment and color was selected (simulating the brick falling down on the conveyour belt) and than 10 images was created while the brick was moved across...
-
Bees Detection on Images: Study of Different Color Models for Neural Networks
PublicationThis paper presents an approach to bee detection in video streams using a neural network classifier. We describe the motivation for our research and the methodology of data acquisition. The main contribution to this work is a comparison of different color models used as an input format for a feedforward convolutional architecture applied to bee detection. The detection process has is based on a neural binary classifier that classifies...
-
Towards bees detection on images: study of different color models for neural networks
PublicationThis paper presents an approach to bee detection in videostreams using a neural network classifier. We describe the motivationfor our research and the methodology of data acquisition. The maincontribution to this work is a comparison of different color models usedas an input format for a feedforward convolutional architecture appliedto bee detection. The detection process has is based on a neural...
-
Training of Deep Learning Models Using Synthetic Datasets
PublicationIn order to solve increasingly complex problems, the complexity of Deep Neural Networks also needs to be constantly increased, and therefore training such networks requires more and more data. Unfortunately, obtaining such massive real world training data to optimize neural networks parameters is a challenging and time-consuming task. To solve this problem, we propose an easy-touse and general approach to training deep learning...
-
Models in spatial development
e-Learning Courses -
Image simulation and annotation for color blinded
PublicationIn this paper methods for image simulation as seen by a color blinded and a method for constructing images of perceived color difference are presented. The work is also focused on the interactive color description of an image contents. As a result, the individuals having problems with color discrimination can identify colors in an image.W artykule prezentowane są metody symulacji kolorów w obrazach postrzeganych przez osoby ze...
-
Multibody models for gait analysis
PublicationThe aim of this study was to create multibody biomechanical models to analyze a normal gait of the human. Proposed models can be used to identify joint moments of the lower limbs during normal gait in the single and double support phases. Applying Newton-Euler formulation, following planar models were developed: 1) a mathematical 6DOF model describing a gait in the sagittal plane of the body for single support phase and double...
-
Computational Complexity and Its Influence on Predictive Capabilities of Machine Learning Models for Concrete Mix Design
PublicationThe design of concrete mixtures is crucial in concrete technology, aiming to produce concrete that meets specific quality and performance criteria. Modern standards require not only strength but also eco-friendliness and production efficiency. Based on the Three Equation Method, conventional mix design methods involve analytical and laboratory procedures but are insufficient for contemporary concrete technology, leading to overengineering...
-
GRAPHICAL MODELS
Journals -
STOCHASTIC MODELS
Journals -
Aleksandra Parteka dr hab. inż.
PeopleAbout me: I am an associate professor and head of doctoral studies at the Faculty of Management and Economics, Gdansk University of Technology (GdanskTech, Poland). I got my MSc degree in Economics from Gdansk University of Technology (2003) and Universita’ Politecnica delle Marche (2005), as well as MA degree in Contemporary European Studies from Sussex University (2006, with distinction). I received my PhD in Economics...
-
A Hyperdense Semantic Domain for Discontinuous Behavior in Physical System Models
PublicationMultiple time models have been proposed for the formalization of hybrid dynamic system behavior. The superdense notion of time is a well-known time model for describing event-based systems where several events can occur simultaneously. Hyperreals provide a domain for defining the semantics of hybrid models that is elegantly aligned with first principles in physics. This paper discusses the value of both time models and shows how...
-
Piotr Paradowski dr
PeopleDr Piotr Paradowski's areas of expertise in quantitative social science methods include truncated and censored models, quantile regressions, survival analysis, panel data models, discrete regressions and qualitative choice models, instrumental variable estimation, and hierarchical modeling. He is also an expert in statistical matching and statistical methods to handle missing data. In addition, he conducts research on income and...
-
Uncertainty estimation of loop impedance measurement determined by the vector method
PublicationThis article presents a detailed analysis of uncertainty estimation of loop impedance measurement determined by the vector method. The analysis includes the following estimates: resistance variance, voltage variance and time measurement variance. This paper presents a methodology for estimating the combined standard uncertainty of loop impedance by the vector method. The vector method allows to determine loop impedance based on...
-
Comparative analysis of different numerical models of a steel radial gate
PublicationHydrotechnical structures are important components in water management system and general flooding safety. Their reliability should be ensured since potential damage might lead to catastrophic consequences. Weir gates are considered to be highly vulnerable elements of each hydro power plant, with regard to its dynamic resistance. The aim of the paper is to compare different numerical models and their influence on the results of...
-
[AEiE] Selected topics of electrical engineering - models of electrical machines
e-Learning Courses{mlang pl} Dyscyplina: automatyka, elektronika i elektrotechnika Zajęcia fakultatywne dla doktorantów II roku Prowadzący: dr hab. inż. Andrzej Wilk, prof. PG, prof. dr hab. inż. Zbigniew Krzemiński Liczba godzin: 15 Forma zajęć: wykład {mlang} {mlang en} Discipline: control, electronic and electrical engineering Facultative course for 2nd-year PhD students Academic teachers: dr hab. inż. Andrzej Wilk, prof. PG, prof....
-
Matrix Strengthening the Identification of Observations with Split Functional Models in the Squared Msplit(q) Estimation Process
PublicationThis article addresses the issue of raising the level of identification of observations with either single or more split functional models in the squared Msplit(q) estimation process. The theoretical part of the study presents the theoretical grounds for the classical method for estimating parameters in a split functional model and proposes a modification of the computational algorithm to increase the quality of the determinations...
-
Modification of Selected Propagation Models in Terms of Path Loss Estimation in Container Terminal
PublicationIt is particularly important to look for any propagation model that could be useful for designing mobile radio networks in container terminal environment. Selected propagation models have been investigated. Firstly - basing on measurements results - they have been evaluated in this scope and the analysis has shown, that the adjustment is needed. This modification improved significantly the accuracy of path loss modelling. For the...
-
Justyna Signerska-Rynkowska dr inż.
PeopleI am currently an assistant professor (adjunct) at Gdansk University of Technology (Department of Differential Equations and Mathematics Applications). My scientific interests include dynamical systems theory, chaos theory and their applications to modeling of biological phenomena, especially to neurosciences. In June 2013 I completed PhD in Mathematics at the Institute of Mathematics of Polish Academy of Sciences (IMPAN) (thesis...
-
Equivalent single-layer models in deformation analysis of laminated multilayered plates
PublicationThe performance of selected Equivalent Single-Layer (ESL) models is evaluated within several classical benchmark tests for linear static analysis of multi-layered plates. The authors elaborated their own Finite Element software based on the first-order shear deformation theory (FOSD) with some modifications incorporated including a correction of the transverse shear stiffness and an application of zig-zag type functions. Seven...
-
On the Structure of Time in Computational Semantics of a Variable-Step Solver for Hybrid Behavior Analysis
PublicationHybrid dynamic systems combine continuous and discrete behavior. Often, computational approaches are employed to derive behaviors that approximate the analytic solution. An important part of this is the approximation of differential equation behavior by numerical integration. The accuracy and computational efficiency of the integration usually depend on the complexity of the method and its implicated approximation errors, especially...
-
Interpretable deep learning approach for classification of breast cancer - a comparative analysis of multiple instance learning models
PublicationBreast cancer is the most frequent female cancer. Its early diagnosis increases the chances of a complete cure for the patient. Suitably designed deep learning algorithms can be an excellent tool for quick screening analysis and support radiologists and oncologists in diagnosing breast cancer.The design of a deep learning-based system for automated breast cancer diagnosis is not easy due to the lack of annotated data, especially...
-
Modeling of medium flow processes in transportation pipelines - the synthesis of their state-space models and the analysis of the mathematical properties of the models for leak detection purposes
PublicationThe dissertation concerns the issue of modeling the pipeline flow process under incompressible and isothermal conditions, with a target application to the leak detection and isolation systems. First, an introduction to the model-based process diagnostics is provided, where its basic terminology, tools, and methods are described. In the following chapter, a review of the state of the art in the field of leak detection and isolation...
-
Estimation of the uncertainty of the AC/AC transducer output voltage
PublicationThe paper presents a methodology for estimating uncertainty of voltage transducers processing function, on the example of CYVS411D07 voltage transducers. The analysis was conducted using the method based on the GUM Guide propagation of uncertainty law and the determination of the coverage factor based on the effective number of degrees of freedom (t-Student distribution). The paper also presents example evaluation results of measured...
-
New results on estimation bandwidth adaptation
PublicationThe problem of identification of a nonstationary autoregressive signal using non-causal estimation schemes is considered. Noncausal estimators can be used in applications that are not time-critical, i.e., do not require real-time processing. A new adaptive estimation bandwidth selection rule based on evaluation of pseudoprediction errors is proposed, allowing one to adjust tracking characteristics of noncausal estimators to unknown...
-
Semantics for an Interdisciplinary Computation
PublicationSemantics for an interdisciplinary computation is becoming increasingly difficult to capture while dealing with multi-domain problems. Expertise from Computer Science, Computer Engineering, Electrical Engineering, and other disciplines merges as engineering challenges in modern systems, such as, Cyber-Physical Systems, Smart Cities, and Bionic Systems must be tackled in a methodological manner. In this paper, a paradigm for formalization...
-
Minimizing Distribution and Data Loading Overheads in Parallel Training of DNN Acoustic Models with Frequent Parameter Averaging
PublicationIn the paper we investigate the performance of parallel deep neural network training with parameter averaging for acoustic modeling in Kaldi, a popular automatic speech recognition toolkit. We describe experiments based on training a recurrent neural network with 4 layers of 800 LSTM hidden states on a 100-hour corpora of annotated Polish speech data. We propose a MPI-based modification of the training program which minimizes the...
-
Melody Harmonization with Interpolated Probabilistic Models
PublicationMost melody harmonization systems use the generative hidden Markov model (HMM), which model the relation between the hidden chords and the observed melody. Relations to other variables, such as the tonality or the metric structure, are handled by training multiple HMMs or are ignored. In this paper, we propose a discriminative means of combining multiple probabilistic models of various musical variables by means of model interpolation....
-
A predictive estimation based control strategy for a quasi-resonant dc-link inverter
PublicationIn this paper the predictive estimation based control strategy for a quasi-resonant dc link inverter (PQRDCLI) is developed. Instead of direct measurement of dc link input inverter current – its estimation with one step prediction is applied. The PQRDCLI fed induction motor, controlled with a predictive current estimation stabilized inverter output voltage slopes independently of load. Moreover, reduction of overvoltage spikes...
-
Disease Models & Mechanisms
Journals -
Kinetic and Related Models
Journals -
Uncertainty analysis of measuring system for instantaneous power research
PublicationThe paper presents a metrological analysis of the measurement system used for diagnosis of induction motor bearings, based on the analysis of the instantaneous power. This system was implemented as a set of devices with dedicated software installed on a PC. A number of measurements for uncertainty estimation was carried out. The results of the measurements are presented in the paper. The results of the aforementioned analysis helped...
-
Language Models in Speech Recognition
PublicationThis chapter describes language models used in speech recognition, It starts by indicating the role and the place of language models in speech recognition. Mesures used to compare language models follow. An overview of n-gram, syntactic, semantic, and neural models is given. It is accompanied by a list of popular software.
-
Identification of models and signals robust to occasional outliers
PublicationIn this paper estimation algorithms derived in the sense of the least sum of absolute errors are considered for the purpose of identification of models and signals. In particular, off-line and approximate on-line estimation schemes discussed in the work are aimed at both assessing the coefficients of discrete-time stationary models and tracking the evolution of time-variant characteristics of monitored signals. What is interesting,...
-
Identification of models and signals robust to occasional outliers
PublicationIn this paper estimation algorithms derived in the sense of the least sum of absolute errors are considered for the purpose of identification of models and signals. In particular, off-line and approximate on-line estimation schemes discussed in the work are aimed at both assessing the coefficients of discrete-time stationary models and tracking the evolution of time-variant characteristics of monitored signals. What is interesting,...
-
Analysis of Interspike-Intervals for the General Class of Integrate-and-Fire Models with Periodic Drive
PublicationWe study one-dimensional integrate-and-fire models of the general type x˙=F (t, x) and analyze properties of the firing map which iterations recover consecutive spike timings. We impose very week constraints for the regularity of the function F (t, x), e.g. often it suffices to assume that F is continuous. If additionally F is periodic in t, using mathematical study of the displacement sequence of an orientation preserving circle...
-
Models of Structures in Didactics
PublicationThe final aim of teaching students subjects, such as structural mechanics, reinforced concrete, and steel structures is to teach them how structures work in a given building as well as to provide them with skills enabling them to calculate and design structures. The behavioral model of the structure, contrary to the architectural model, which focuses mainly on the external form of the building, shows workings from both the static...
-
Bridging the gap between business process models and use-case models
PublicationToday's software development methodologies are equipped with a plethora of methods and techniques for business process engineering and Requirements Engineering. However, heavy investments in IT have not brought forth expected results. What seems to be lacking is a systematic approach that consolidates both disciplines to gain a synergistic effect. To address this challenge we extend Use-Case Driven Approach (UCDA) by binding use...
-
Equivalent Single Layer Models in Free Vibration Analysis of Laminated Multi-Layered Plates
PublicationThe performance of selected equivalent single-layer (ESL) models is evaluated within several classical benchmark tests for small amplitude free vibration analysis of multi-layered plates. The authors elaborated their own Finite Element software based on the first-order shear deformation (FOSD) theory with some modifications incorporated including a correction of the transverse shear stiffness and an application of zigzag type functions....
-
Comparison of technology adoption models
PublicationThere are several technology adoption models, that try to explain, how and why the technologies are adopted and used. Among those, that are widely used to explain, how the older adults accept technologies, there are some general models and models specific to the group of older users. Among the general ones I would recommend paying attention to the following models: Technology Acceptance Model (TAM) proposed by Davis...
-
Kriging Models for Microwave Filters
PublicationSurrogate modeling of microwave filters’ response is discussed. In particular, kriging is used to model either the scattering parameters of the filter or the rational representation of the filter’s characteristics. Surrogate models for these two variants of kriging are validated in solving a microwave filter optimization problem. A clear advantage of surrogate models based on the rational representation over the models based on scattering...
-
A new method of presosns identification based on comparative analysis of 3D face models
PublicationThe article presents the use of modern close range photogrammetry for possessing highly accurate 3D models of the human face (including the ears). Modern methods used to obtain precise data describing the construction of a human face, and even the whole human body, should allow to get finished measurement material in a very short time. Those features belong to the optical scanning technology. Comparative analysis of models of the...
-
COMPARISON OF INFINITE ELEMENT MODELS
PublicationThe main objective of this paper is to show the comparison of two models of infinite ab- sorbing layer with increasing damping in numerical investigations of elastic wave prop- agation in unbounded structures. This has been achieved by the Authors by a careful in- vestigation of two different engineering structures characterised by gradually increasing geometrical and mathematical description complexities. The analysis included...
-
Comparison of soil models in the thermodynamic analysis of a submarine pipeline buried in seabed sediments
PublicationThis paper deals with mathematical modelling of a seabed layer in the thermodynamic analysis of a submarine pipeline buried in seabed sediments. The existing seabed soil models: a “soil ring” and a semi-infinite soil layer are discussed in a comparative analysis of the shape factor of a surrounding soil layer. The meaning of differences in the heat transfer coefficient of a soil layer is illustrated based on a computational example...
-
Estimation of the uncertainty of the LEM CV 3-500 transducers processing function
PublicationThe paper presents the methodology for estimating uncertainty of a voltage transducer processing function based on the example of LEM CV 3-500 transducer. The analysis was performed using the method based on the GUM Guide propagation of uncertainty law and the Monte Carlo numerical method. The article also presents a comparative analysis of the results of a voltage measurement evaluation done using the Monte Carlo simulation method,...
-
Non-monotone graph searching models
PublicationGraph searching encompasses a variety of different models, many of which share a property that in optimal strategies fugitive can never access once searched regions. Monotonicity, as it is called, is vital in many established results in the field however its absence significantly impedes the analysis of a given problem. This survey attempts to gather non-monotone models, that are less researched in effort of summarizing the results...