Search
Filters
total: 871
Search results for: QUANTIZATION AWARE TRAINING
-
Conference on Software Engineering Education and Training (Conference on Software Engineering Education and Training (previously Conference is Software Engineering Education, CSEE, changed in 1997))
Conferences -
Tinnitus Therapy Based on High-Frequency Linearization
PublicationThe aim of this work was to present problems related to tinnitus symptoms, its pathogenesis, hypotheses on tinnitus causes, and therapy treatments to reduce or mask the phantom noise. In addition, the hypothesis on the existence of parasitic quantization that accompanies hearing loss was recalled. The paper contains a description of experiments carried out with the application of high-frequency dither having specially formed spectral...
-
Piotr Lorens prof. dr hab. inż. arch.
PeopleProf. Piotr Lorens - PhD, DSc., urban planner. Lecturer in urban design and development and - since 2007 - Head of the Department of Urban Design and Regional Planning at the Faculty of Architecture, Gdansk University of Technology. His university activities also include coordination of the research and EU Social Fund projects. Based on these projects he was responsible for organization and coordination of the post-graduate studies...
-
Quantum corrections to phi^4 model solutions and applications to Heisenberg chain dynamics
PublicationThe Heisenberg spin chain is considered in φ^4 model approximation. Quantum corrections to classical solutions of the one-dimensional φ^4 model within the correspondent physics are evaluated with account of rest d−1 dimensions of a d-dimensional theory. A quantization of the model is considered in terms of spacetime functional integral. The generalized zeta-function formalism is used to renormalize and evaluate the functional integral...
-
Quantum corrections to 4 model solutions and applications to Heisenberg chain dynamics
PublicationThe Heisenberg spin chain is considered in φ^4 model approximation. Quantum corrections to classical solutions of the one-dimensional φ^4 model within the correspondent physics are valuated with account of rest d − 1 dimensions of a d-dimensional theory. A quantization of the model is considered in terms of space- time functional integral. The generalized zeta-function formalism is used to renormalize and evaluate the functional...
-
Ontology Oriented Approach to Service Selection and Invocation in Complex Context Analysis
PublicationContext-aware applications running in the intelligent space are taken into account and their execution in the service oriented environment is considered. It has been presented where and how SOA services can be utilized during their execution: to analyze current context of the application and to support execution of strictly determined actions suitable for that context. The proposed mechanism of context-aware service selection and...
-
Leszek Pawłowski dr
PeopleDr Leszek Pawłowski, PhD, lawyer, assistant professor at the Department of Palliative Medicine, Medical University of Gdańsk (Poland), member of Polish Working Group on End-of-Life Ethics and Editorial Board member of the journal Palliative Medicine in Practice. His research is focused on legal issues in palliative care, hospice and palliative care volunteering, Advance Care Planning and medical law. He is involved in healthcare...
-
APPLICATION OF THE HIGH FREQUENCY LINEARIZATION OF THE EAR IN PATIENTS WITH TINNITUS . Metoda linearyzacji narządu słuchu u osób cierpiących z szumami usznymi
PublicationThis paper summarises the problem of tinnitus, hypotheses on its causes and the treatment methods. Moreover, a hypothesis on tinnitus origins is explained, based on the mechanisms of the analog-to-digital conversion and quantization. In addition, this paper describes methods of determining the acoustic intensity and spectra of low- level ultrasonic signals, as well as impedance characteristics of an ultrasound transducer. Furthermore,...
-
Multiscaled Hybrid Features Generation for AdaBoost Object Detection
PublicationThis work presents the multiscaled version of modified census features in graphical objects detection with AdaBoost cascade training algorithm. Several experiments with face detector training process demonstrate better performance of such features over ordinal census and Haar-like approaches. The possibilities to join multiscaled census and Haar features in single hybrid cascade of strong classifiers are also elaborated and tested....
-
Affect-awareness framework for intelligent tutoring systems
PublicationThe paper proposes a framework for construction of Intelligent Tutoring Systems (ITS), that take into consideration student emotional states and make affective interventions. The paper provides definitions of `affect-aware systems' and `affective interventions' and describes the concept of the affect-awareness framework. The proposed framework separates emotion recognition from its definition, processing and making decisions on...
-
Serendipitous Recommendations Through Ontology-Based Contextual Pre-filtering
PublicationContext-aware Recommender Systems aim to provide users with better recommendations for their current situation. Although evaluations of recommender systems often focus on accuracy, it is not the only important aspect. Often recommendations are overspecialized, i.e. all of the same kind. To deal with this problem, other properties can be considered, such as serendipity. In this paper, we study how an ontology-based and context-aware...
-
Biometric identity verification
PublicationThis chapter discusses methods which are capable of protecting automatic speaker verification systems (ASV) from playback attacks. Additionally, it presents a new approach, which uses computer vision techniques, such as the texture feature extraction based on Local Ternary Patterns (LTP), to identify spoofed recordings. We show that in this case training the system with large amounts of spectrogram patches may be difficult, and...
-
Development of advanced machine learning for prognostic analysis of drying parameters for banana slices using indirect solar dryer
PublicationIn this study, eXtreme Gradient Boosting (XGBoost) and Light Gradient Boosting (LightGBM) algorithms were used to model-predict the drying characteristics of banana slices with an indirect solar drier. The relationships between independent variables (temperature, moisture, product type, water flow rate, and mass of product) and dependent variables (energy consumption and size reduction) were established. For energy consumption,...
-
Mykola Lukianov mgr
PeopleMykola Lukianov received a B.Sc. and M.Sc degree in Electronics and Communications from the National Technical University of Ukraine “Igor Sikorsky Kyiv Polytechnic Institute” in 2018 and 2020 respectively. He is currently a PhD student at the Gdansk University of Technology, Poland and a researcher group member in project SMARTGYsum “Research and Training Network for Smart and Green Energy Systems and Business Models”. His research...
-
SASD Support for training and career development of researchers
ProjectsProject realized in Faculty of Management and Economics according to 247549 agreement from 2010-09-10
-
Quantum Coherence as a Resource
PublicationThe coherent superposition of states, in combination with the quantization of observables, represents one of the most fundamental features that mark the departure of quantum mechanics from the classical realm. Quantum coherence in many-body systems embodies the essence of entanglement and is an essential ingredient for a plethora of physical phenomena in quantum optics, quantum information, solid state physics, and nanoscale thermodynamics....
-
Investigation into MPI All-Reduce Performance in a Distributed Cluster with Consideration of Imbalanced Process Arrival Patterns
PublicationThe paper presents an evaluation of all-reduce collective MPI algorithms for an environment based on a geographically-distributed compute cluster. The testbed was split into two sites: CI TASK in Gdansk University of Technology and ICM in University of Warsaw, located about 300 km from each other, both connected by a fast optical fiber Ethernet-based 100 Gbps network (900 km part of the PIONIER backbone). Each site hosted a set...
-
A CONTEXT IN RECOMMENDER SYSTEMS
PublicationRecommender systems aim to propose potentially interesting items to a user based on his preferences or previous interaction with the system. In the last decade, researcher found out that known recommendation techniques are not sufficient to predict user decisions. It has been noticed that user preferences strongly depend on the context in which he currently is. This raises new challenges for the researchers such as how to obtain...
-
Food Classification from Images Using a Neural Network Based Approach with NVIDIA Volta and Pascal GPUs
PublicationIn the paper we investigate the problem of food classification from images, for the Food-101 dataset extended with 31 additional food classes from Polish cuisine. We adopted transfer learning and firstly measured training times for models such as MobileNet, MobileNetV2, ResNet50, ResNet50V2, ResNet101, ResNet101V2, InceptionV3, InceptionResNetV2, Xception, NasNetMobile and DenseNet, for systems with NVIDIA Tesla V100 (Volta) and...
-
Active Learning Based on Crowdsourced Data
PublicationThe paper proposes a crowdsourcing-based approach for annotated data acquisition and means to support Active Learning training approach. In the proposed solution, aimed at data engineers, the knowledge of the crowd serves as an oracle that is able to judge whether the given sample is informative or not. The proposed solution reduces the amount of work needed to annotate large sets of data. Furthermore, it allows a perpetual increase...
-
Texture Features for the Detection of Playback Attacks: Towards a Robust Solution
PublicationThis paper describes the new version of a method that is capable of protecting automatic speaker verification (ASV) systems from playback attacks. The presented approach uses computer vision techniques, such as the texture feature extraction based on Local Ternary Patterns (LTP), to identify spoofed recordings. Our goal is to make the algorithm independent from the contents of the training set as much as possible; we look for the...
-
Process of Medical Dataset Construction for Machine Learning-Multifield Study and Guidelines
PublicationThe acquisition of high-quality data and annotations is essential for the training of efficient machine learning algorithms, while being an expensive and time-consuming process. Although the process of data processing and training and testing of machine learning models is well studied and considered in the literature, the actual procedures of obtaining data and their annotations in collaboration with physicians are in most cases...
-
Massive Open Online Courses (MOOCs) in hospitality and tourism
PublicationThe tourism industry, interesting and challenging, faces structural human resource problems such as skills shortages and staff turnover, seasonality and a high percentage of small to medium enterprises whose employees have limited time for training or education. Large tourism enterprises often span countries and continents, such as hotel chains, airlines, cruise companies and car rentals, where the employees need similar training...
-
Feature Reduction Using Similarity Measure in Object Detector Learning with Haar-like Features
PublicationThis paper presents two methods of training complexity reduction by additional selection of features to check in object detector training task by AdaBoost training algorithm. In the first method, the features with weak performance at first weak classifier building process are reduced based on a list of features sorted by minimum weighted error. In the second method the feature similarity measures are used to throw away that features...
-
Tinnitus Therapy Based on High-Frequency Linearization Principles - Preliminary Results
PublicationThe aim of this work is to present problems related to tinnitus symptoms, its pathogenesis, hypotheses on tinnitus causes, and therapy treatment to reduce or mask the phantom noise. In addition, the hypothesis on the existence of parasitic quantization that accompanies hearing loss has been recalled. Moreover, the paper describes a study carried out by the Authors with the application of high-frequency dither having specially formed...
-
Top k Recommendations using Contextual Conditional Preferences Model
PublicationRecommender systems are software tools and techniques which aim at suggesting to users items they might be interested in. Context-aware recommender systems are a particular category of recommender systems which exploit contextual information to provide more adequate recommendations. However, recommendation engines still suffer from the cold-start problem, namely where not enough information about users and their ratings is available....
-
Performance improvement of NN based RTLS by customization of NN structure - heuristic approach
PublicationThe purpose of this research is to improve performance of the Hybrid Scene Analysis – Neural Network indoor localization algorithm applied in Real-time Locating System, RTLS. A properly customized structure of Neural Network and training algorithms for specific operating environment will enhance the system’s performance in terms of localization accuracy and precision. Due to nonlinearity and model complexity, a heuristic analysis...
-
Computer-Supported Polysensory Integration Technology for Educationally Handicapped Pupils
PublicationIn this paper, a multimedia system providing technology for hearing and visual attention stimulation is shortly presented. The system aims to support the development of educationally handicapped pupils. The system has been presented in the context of its configuration, architecture, and therapeutic exercise implementation issues. Results of pupils’ improvements after 8 weeks of training with the system are also provided. Training...
-
Adaptive CAD-Model Construction Schemes
PublicationTwo advanced surrogate model construction techniques are discussed in this paper. The models employ radial basis function (RBF)interpolation scheme or artificial neural networks (ANN) with a new training algorithm. Adaptive sampling technique is applied withrespect to all variables. Histograms showing the quality of the models are presented. While the quality of RBF models is satisfactory, theperformance of the ANN models obtained...
-
Paweł Śliwiński dr hab. inż.
PeopleDiplomas and academic degrees. Training 2017: degree of habilitated doctor;2006: PhD in Technical Sciences. PhD thesis defended with distinction. Gdansk University of Technology, Faculty of Mechanical Engineering.2000: Master of Science. Graduated from the university with distinction. Gdansk University of Technology, Faculty of Mechanical Engineering.1995: Mechanical technician. Graduated from a Technical Secondary School...
-
Natalia Sokół dr inż.
PeopleBACKGROUND Master of Science in Light and Lighting (2008-2009/11) The UCL Bartlett School of Graduate Studies, Faculty of the Built Environment, London, UK, www.bartlett.ucl.ac.uk MA Degree in Interior Architecture (1999-2004), The Academy of Fine Arts, Poznan, Poland, www.uap.edu.pl MA Degree in Art Education (1997-2002), Academy of Fine Arts, Poznan, Poland, www.uap.edu.pl MAIN RESEARCH AREAS · ...
-
Identification of the Contamination Source Location in the Drinking Water Distribution System Based on the Neural Network Classifier
PublicationThe contamination ingression to the Water Distribution System (WDS) may have a major impact on the drinking water consumers health. In the case of the WDS contamination the data from the water quality sensors may be efficiently used for the appropriate disaster management. In this paper the methodology based on the Learning Vector Quantization (LVQ) neural network classifier for the identification of the contamination source location...
-
A review of emotion recognition methods based on keystroke dynamics and mouse movements
PublicationThe paper describes the approach based on using standard input devices, such as keyboard and mouse, as sources of data for the recognition of users’ emotional states. A number of systems applying this idea have been presented focusing on three categories of research problems, i.e. collecting and labeling training data, extracting features and training classifiers of emotions. Moreover the advantages and examples of combining standard...
-
Dominika Wróblewska dr inż. arch.
PeopleDr. Eng. arch. Dominika Wróblewska, university professor, obtained the title of doctor of technical sciences in 2000. In 2002, she started working at the Faculty of Hydro and Environmental Engineering at the Gdańsk University of Technology (currently the Faculty of Civil and Environmental Engineering) as an assistant professor. Since 2019, he has been working as a university professor. The areas of interest are changes, introducing...
-
Izabela Mironowicz dr hab. inż. arch.
People -
Recognizing emotions on the basis of keystroke dynamics
PublicationThe article describes a research on recognizing emotional states on the basis of keystroke dynamics. An overview of various studies and applications of emotion recognition based on data coming from keyboard is presented. Then, the idea of an experiment is presented, i.e. the way of collecting and labeling training data, extracting features and finally training classifiers. Different classification approaches are proposed to be...
-
Beata Krawczyk-Bryłka dr
PeoplePsycholog, doktor nauk humanistycznych w dziedzinie zarządzania, adiunkt w Katedrze przedsiębiorczości. 2018 - 2021: Kierownik projektu NCN: „Efektuacyjny model zespołu przedsiębiorczego. Jak działają przedsiębiorcze zespoły odnoszące sukces" od 2016: Quality Standards Lead filaru People management & personal development na studiach MBA Politechniki Gdańskiej 2008 – 2012: Prodziekan ds kształcenia Wzydziału Zarządzania i Ekonomii...
-
Musical Instrument Identification Using Deep Learning Approach
PublicationThe work aims to propose a novel approach for automatically identifying all instruments present in an audio excerpt using sets of individual convolutional neural networks (CNNs) per tested instrument. The paper starts with a review of tasks related to musical instrument identification. It focuses on tasks performed, input type, algorithms employed, and metrics used. The paper starts with the background presentation, i.e., metadata...
-
Color-based Detection of Bleeding in Endoscopic Images
PublicationIn this paper a color descriptor designed for bleeding detection in endoscopic images is proposed. The development of the algorithm was carried out on a representative training set of 36 images of bleeding and 25 clear images. Another 38 bleeding and 26 normal images were used in the final stage as a test set. All of the considered images were extracted from separate endoscopic examinations. The experiments include color distribution...
-
Paweł Ziemiański dr
PeoplePaweł Ziemiański - assistant professor at the Faculty of Management and Economics of the Gdańsk University of Technology. His research interests concern entrepreneurship, entrepreneurial intentions, and the dark side of entrepreneurship. He also conducted research regarding psychological aspects of exercising power in organizations. In teaching, he is interested in developing and using case studies. He took part in the training...
-
Simple Measurement Method for Resistive Sensors Based on ADCs of Microcontrollers
PublicationA new, complete measurement method for resistance measurement of resistive sensors for systems based on microcontrollers equipped with analog-to-digital converters (ADCs) is proposed. The interface circuit consists of only four resistors, including a resistive sensor and a reference resistor, connected directly to the microcontroller pins. It is activated only during measurements, which significantly reduces power consumption....
-
Affective reactions to playing digital games
PublicationThe paper presents a study of emotional states during a gameplay. An experiment of two-player Tetris game is reported, followed by the analysis of the results - self-reported emotional states as well as physiological signals measurements interpretation. The study reveals the diversity of emotional reactions and concludes, that a representative player's emotional model is hard to define. Instead, an adaptive approach to emotion...
-
Towards Scalable Simulation of Federated Learning
PublicationFederated learning (FL) allows to train models on decentralized data while maintaining data privacy, which unlocks the availability of large and diverse datasets for many practical applications. The ongoing development of aggregation algorithms, distribution architectures and software implementations aims for enabling federated setups employing thousands of distributed devices, selected from millions. Since the availability of...
-
Poprawa jakości klasyfikacji głębokich sieci neuronowych poprzez optymalizację ich struktury i dwuetapowy proces uczenia
PublicationW pracy doktorskiej podjęto problem realizacji algorytmów głębokiego uczenia w warunkach deficytu danych uczących. Głównym celem było opracowanie podejścia optymalizującego strukturę sieci neuronowej oraz zastosowanie uczeniu dwuetapowym, w celu uzyskania mniejszych struktur, zachowując przy tym dokładności. Proponowane rozwiązania poddano testom na zadaniu klasyfikacji znamion skórnych na znamiona złośliwe i łagodne. W pierwszym...
-
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...
-
Model of emotions for game players
PublicationAffect-aware video games can respond to a game player's emotions. Such games seem to be more attractive for users. Therefore for that kind of games it is necessary to create a model of the player's emotions to know to which emotions the application should react. The paper describes different models of emotions. The questionnaire and experiment for video game players is presented. Some results of the tests are shown. Then the model...
-
Polski PR-owiec w świecie ponowoczesnym – wyniki badań i ich analiza. Poza dobrem i złem
PublicationW artykule analizuję wyniki badania Profesjonalizm public relations w Polsce, przeprowadzonego na potrzeby budowanego dokumentu Standardy profesjonalnego public relations – prezentowanego w tej pracy. Stawiam również pytanie o przyczynę częstych odpowiedzi respondentów „czasami” czy „nie mam zdania” i wykazuję niespójność systemów wartości respondentów zestawiając ich odpowiedzi na poszczególne pytania. Stawiam tezę, że niespójności...
-
Spatial Modelling in Environmental Analysis and Civil Engineering
PublicationAs can be seen, Spatial Modelling in Environmental Analysis and Civil Engineering applies to a multitude of applications. In an attempt to bring the topic closer to potential readers, the special issue referred to security methods, optimization of calculations, conducting measurements and empirical tests. Each article has made an original contribution to the development of practical science that can be replicated by researchers...
-
Train the trainer course
PublicationThis chapter presents the concept, evaluation and evaluation results for the train the trainer. This concept of train the trainers is prepared within Workpackage 5 of EU-funded project: MASTER BSR (Erasmus+ Strategic Partnership Programme). Due to the nature of adult learning the content is designed for the use of participatory methods (involved, active). This method uses various techniques of active learning e.g. group work,...
-
Józef Woźniak prof. dr hab. inż.
PeopleProfessor Józef Woźniak received his M.Sc., Ph.D. and D.Sc. degrees in electronics and telecommunications from the Faculty of Electronics, Gdańsk University of Technology (GUT), Poland, in 1971, 1976 and 1991, respectively. In January 2002 he became a full professor. In 1993 he was elected Associate Dean of the Faculty of Electronics and in 1994 he was he was appointed a Professor at GUT. Simultaneously, from October 1994 till...