Search
Filters
total: 862
Search results for: quantization aware training
-
Deep Learning Optimization for Edge Devices: Analysis of Training Quantization Parameters
PublicationThis paper focuses on convolution neural network quantization problem. The quantization has a distinct stage of data conversion from floating-point into integer-point numbers. In general, the process of quantization is associated with the reduction of the matrix dimension via limited precision of the numbers. However, the training and inference stages of deep learning neural network are limited by the space of the memory and a...
-
Performance and Energy Aware Training of a Deep Neural Network in a Multi-GPU Environment with Power Capping
PublicationIn this paper we demonstrate that it is possible to obtain considerable improvement of performance and energy aware metrics for training of deep neural networks using a modern parallel multi-GPU system, by enforcing selected, non-default power caps on the GPUs. We measure the power and energy consumption of the whole node using a professional, certified hardware power meter. For a high performance workstation with 8 GPUs, we were...
-
Emotion Recognition and Its Applications
PublicationThe paper proposes a set of research scenarios to be applied in four domains: software engineering, website customization, education and gaming. The goal of applying the scenarios is to assess the possibility of using emotion recognition methods in these areas. It also points out the problems of defining sets of emotions to be recognized in different applications, representing the defined emotional states, gathering the data and...
-
Musical Instrument Tagging Using Data Augmentation and Effective Noisy Data Processing
PublicationDeveloping signal processing methods to extract information automatically has potential in several applications, for example searching for multimedia based on its audio content, making context-aware mobile applications (e.g., tuning apps), or pre-processing for an automatic mixing system. However, the last-mentioned application needs a significant amount of research to reliably recognize real musical instruments in recordings....
-
Exploring the influence of personal factors on physiological responses to mental imagery in sport
PublicationImagery is a well-known technique in mental training which improves performance efficiency and influences physiological arousal. One of the biomarkers indicating the amount of physiological arousal is skin conductance level (SCL). The aim of our study is to understand how individual differences in personality (e.g. neuroticism), general imagery and situational sport anxiety are linked to arousal measuring with SCL in situational...
-
Affect aware video games
PublicationIn this chapter a problem of affect aware video games is described, including such issue as: emotional model of the player, design, development and UX testing of affect-aware video games, multimodal emotion recognition and a featured review of affect-aware video games.
-
Modeling emotions for affect-aware applications
PublicationThe chapter concerns emotional states representation and modeling for software systems, that deal with human affect. A review of emotion representation models is provided, including discrete, dimensional and componential models. The paper provides also analysis of emotion models used in diverse types of affect-aware applications: games, mood trackers or tutoring systems. The analysis is supported with two design cases. The study...
-
Design Elements of Affect Aware Video Games
PublicationIn this paper issues of design and development process of affect-aware video games are presented. Several important design aspects of such games are pointed out. A concept of a middleware framework is proposed that separates the development of affect-aware video games from emotion recognition algorithms and support from input sensors. Finally, two prototype affect-aware video games are presented that conform to the presented architecture...
-
Emotion Recognition for Affect Aware Video Games
PublicationIn this paper the idea of affect aware video games is presented. A brief review of automatic multimodal affect recognition of facial expressions and emotions is given. The first result of emotions recognition using depth data as well as prototype affect aware video game are presented
-
Methods of Cyclist Training in Europe
PublicationThe following study aims to address the issue of cyclist training methodologies. Recent European bicycle accident statistics reveal a troubling upward trend. A potential solution to mitigate such incidents involves providing cyclists with comprehensive training encompassing traffic regulations and interactions with fellow road users. We conducted a comparative analysis of the cycling education approaches and cyclist training systems...
-
Regularization as quantization in reducible representations of CCR
PublicationOpis kwantowego pola elektromagnetycznego przy pomocy redukowalnych reprezentacji CCR prowadzi do automatycznej regularyzacji teorii. Sformułowanie jest jawnie relatywistycznie współzmiennicze. Przeanalizowano - jako przykład - pola kwantowe wytwarzane przez klasyczne źródła.
-
States of light via reducible quantization.
PublicationRelatywistyczne sformułowanie kwantowania pola opartego o redukowalne reprezentacje kanonicznych związków komutacyjnych. Konstrukcja stanów fokowskich i koherentnych. Analiza automatycznej regularyzacji rozbieżności w podczerwieni. Twierdzenie o granicy termodynamicznej.
-
Geometry, Integrability and Quantization
Journals -
Using Different Information Channels for Affect-Aware Video Games - A Case Study
PublicationThis paper presents the problem of creating affect-aware video games that use different information channels, such as image, video, physiological signals, input devices, and player’s behaviour, for emotion recognition. Presented case studies of three affect-aware games show certain conditions and limitations for using specific signals to recognize emotions and lead to interesting conclusions.
-
Evaluation of affective intervention process in development of affect-aware educational video games
PublicationIn this paper initial experiences are presented on implementing specific methodology of affective intervention design (AFFINT) for development of affect-aware educational video games. In the described experiment, 10 student teams are to develop affect-aware educational video games using AFFINT to formalize the whole process. Although all projects are still in progress, first observations and conclusions may already be presented.
-
Design and Configuration of Context-aware VoIP Telephony Systems
PublicationVoice over IP is a widely used concept with regard to a realization technology of different types of telephony systems, including those that are used in enterprises. Such systems of a call procesing component and a set of desk endpoints that are pervasove from a user perspective. Those andpoints are usually not mobile, but in result can deliver a much greater set of functions needed in an everyday office work. Those functions also...
-
Health and safety training for students
e-Learning CoursesHealth and safety training is aimed at all newcoming students of Gdańsk University of Technology.
-
Positively aware : the monthly journal of the Test Positive Aware Network
Journals -
EvOLAP Graph – Evolution and OLAP-Aware Graph Data Model
PublicationThe objective of this paper is to propose a graph model that would be suitable for providing OLAP features on graph databases. The included features allow for a multidimensional and multilevel view on data and support analytical queries on operational and historical graph data. In contrast to many existing approaches tailored for static graphs, the paper addresses the issue for the changing graph schema. The model, named Evolution...
-
eNauczanie PG - introductory training for students
e-Learning CoursesThe eNauczanie platform, an introductory training for foreign students.
-
Multi-domain and Context-Aware Recommendations Using Contextual Ontological User Profile
PublicationRecommender Systems (RS) became popular tools in many Web services like Netflix, Amazon, or YouTube, because they help a~user to avoid an information overload problem. One of the types of RS are Context-Aware RS (CARS) which exploit contextual information to provide more adequate recommendations. Cross-Domain RS (CDRS) were created as a response to the data sparsity problem which occurs when only few users can provide reviews or...
-
Reducible representations of CAR and CCR with possible applications to field quantization.
PublicationRedukowalne reprezentacje CAR i CCR zastosowane są do drugiej kwantyzacji pól Diraca i Maxwella. Powstające w ten sposób operatory pola są rzeczywiście operatorami, a nie dystrybucjami o wartościach operatorowych. Przykłady pokazują, że formalizm taki może prowadzić do skończonej teorii pola.
-
Rendezvous of Distance-Aware Mobile Agents in Unknown Graphs
PublicationWe study the problem of rendezvous of two mobile agents starting at distinct locations in an unknown graph. The agents have distinct labels and walk in synchronous steps. However the graph is unlabelled and the agents have no means of marking the nodes of the graph and cannot communicate with or see each other until they meet at a node. When the graph is very large we want the time to rendezvous to be independent of the graph size...
-
Baltic Phytoremediation - Training Module
e-Learning CoursesThis is an on-line training course developed in BAPR - Baltic Phytoremediation project, co-financed by Interreg South Baltic Programme 2014 - 2020. Here, You can learn about the basics and potential of phytoremediation to clean soil and see the real-life applications of this technology. In order to be enrolled to this course, you’ll need to create a local account with the user ID provided. When your account is generated, you...
-
CIP Security Awareness and Training: Standards and Practice
PublicationThese are critical infrastructure employees who have access to the critical cyber assets in the first place. This situation is well recognised by international and national standardisation bodies which recommend security education, training and awareness as one of the key elements of critical infrastructure protection. In this chapter the standards are identified and their relevant areas are described. A practical implementation...
-
CIP Security Awareness and Training: Standards and Practice
PublicationThese are critical infrastructure employees who have access to the critical cyber assets in the first place. This situation is well recognized by international and national standardization bodies which recommend security education, training and awareness as one of the key elements of critical infrastructure protection. In this chapter the standards are identified and their relevant areas are described. A practical implementation...
-
eNauczanie - introductory training for students - 2020/2021
e-Learning CourseseNauczanie - introductory training for students - 2020/2021
-
Energy-Aware Scheduling for High-Performance Computing Systems: A Survey
PublicationHigh-performance computing (HPC), according to its name, is traditionally oriented toward performance, especially the execution time and scalability of the computations. However, due to the high cost and environmental issues, energy consumption has already become a very important factor that needs to be considered. The paper presents a survey of energy-aware scheduling methods used in a modern HPC environment, starting with the...
-
Neural network training with limited precision and asymmetric exponent
PublicationAlong with an extremely increasing number of mobile devices, sensors and other smart utilities, an unprecedented growth of data can be observed in today’s world. In order to address multiple challenges facing the big data domain, machine learning techniques are often leveraged for data analysis, filtering and classification. Wide usage of artificial intelligence with large amounts of data creates growing demand not only for storage...
-
Exploration of Creativity Techniques in Software Engineering in Training-Application-Feedback Cycle
PublicationCreativity research has proposed about a hundred and fifty creativity techniques. The question is whether they can be applied in software engineering for creativity training or directing creativity in software projects. This paper aims at answering this question via a quasi-experiment conducted in Training-Application-Feedback cycle in which participants express their opinions about selected creativity techniques after training...
-
Process arrival pattern aware algorithms for acceleration of scatter and gather operations
PublicationImbalanced process arrival patterns (PAPs) are ubiquitous in many parallel and distributed systems, especially in HPC ones. The collective operations, e.g. in MPI, are designed for equal process arrival times (PATs), and are not optimized for deviations in their appearance. We propose eight new PAP-aware algorithms for the scatter and gather operations. They are binomial or linear tree adaptations introducing additional process...
-
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...
-
Factors Affecting the Effectiveness of Military Training in Virtual Reality Environment
PublicationIn this paper, we explored the factors influencing the effectiveness of military trainings performed in a virtual reality environment. The rationale for taking up the topic is the fact that such trainings are often conducted under specific operational procedures. These procedures may create rigorous frameworks for all elements of the learning environment, including the teacher’s performance. Therefore, to ensure the most conducive...
-
Energy-Aware High-Performance Computing: Survey of State-of-the-Art Tools, Techniques, and Environments
PublicationThe paper presents state of the art of energy-aware high-performance computing (HPC), in particular identification and classification of approaches by system and device types, optimization metrics, and energy/power control methods. System types include single device, clusters, grids, and clouds while considered device types include CPUs, GPUs, multiprocessor, and hybrid systems. Optimization goals include various combinations of...
-
Training Services in Small and Medium-sized Enterprises: Evidence from Poland
PublicationIn the knowledge-based economy knowledge and skills are becoming more and more significant for the success of companies. This applies also to firms from small and medium-sized enterprises (SMEs) sector. As large companies in many cases posses special divisions devoted to trainings, they normally have no problems with updating the knowledge and skills of their employees. The situation is different with regard to SMEs, which often...
-
Resource constrained neural network training
PublicationModern applications of neural-network-based AI solutions tend to move from datacenter backends to low-power edge devices. Environmental, computational, and power constraints are inevitable consequences of such a shift. Limiting the bit count of neural network parameters proved to be a valid technique for speeding up and increasing efficiency of the inference process. Hence, it is understandable that a similar approach is gaining...
-
Network-aware Data Prefetching Optimization of Computations in a Heterogeneous HPC Framework
PublicationRapid development of diverse computer architectures and hardware accelerators caused that designing parallel systems faces new problems resulting from their heterogeneity. Our implementation of a parallel system called KernelHive allows to efficiently run applications in a heterogeneous environment consisting of multiple collections of nodes with different types of computing devices. The execution engine of the system is open for...
-
Context-Aware Indexing and Retrieval for Cognitive Systems Using SOEKS and DDNA
PublicationVisual content searching, browsing and retrieval tools have been a focus area of interest as they are required by systems from many different domains. Context-based, Content-Based, and Semantic-based are different approaches utilized for indexing/retrieving, but have their drawbacks when applied to systems that aim to mimic the human capabilities. Such systems, also known as Cognitive Systems, are still limited in terms of processing...
-
CMGNet: Context-aware middle-layer guidance network for salient object detection
PublicationSalient object detection (SOD) is a critical task in computer vision that involves accurately identifying and segmenting visually significant objects in an image. To address the challenges of gridding issues and feature...
-
Real and Virtual Instruments in Machine Learning – Training and Comparison of Classification Results
PublicationThe continuous growth of the computing power of processors, as well as the fact that computational clusters can be created from combined machines, allows for increasing the complexity of algorithms that can be trained. The process, however, requires expanding the basis of the training sets. One of the main obstacles in music classification is the lack of high-quality, real-life recording database for every instrument with a variety...
-
Cavity-QED tests of representations of canonical commutation relations employed in field quantization
PublicationDane eksperymentalne dotyczące oscylacji Rabiego porównano z opisem teoretycznym w ramach alternatywnych sformułowań elektrodynamiki kwantowej. Okazało się, iż eksperyment nie jest w stanie rozróżnić opisu standardowego od nowego sformułowania opartego o redukowalne reprezentacje CCR. Zaproponowano nowy eksperyment, którego wynik mógłby być rozstrzygający.
-
Vehicle detector training with minimal supervision
PublicationRecently many efficient object detectors based on convolutional neural networks (CNN) have been developed and they achieved impressive performance on many computer vision tasks. However, in order to achieve practical results, CNNs require really large annotated datasets for training. While many such databases are available, many of them can only be used for research purposes. Also some problems exist where such datasets are not...
-
Improving css-KNN Classification Performance by Shifts in Training Data
PublicationThis paper presents a new approach to improve the performance of a css-k-NN classifier for categorization of text documents. The css-k-NN classifier (i.e., a threshold-based variation of a standard k-NN classifier we proposed in [1]) is a lazy-learning instance-based classifier. It does not have parameters associated with features and/or classes of objects, that would be optimized during off-line learning. In this paper we propose...
-
Vocational training
Journals -
Benchmarking Deep Neural Network Training Using Multi- and Many-Core Processors
PublicationIn the paper we provide thorough benchmarking of deep neural network (DNN) training on modern multi- and many-core Intel processors in order to assess performance differences for various deep learning as well as parallel computing parameters. We present performance of DNN training for Alexnet, Googlenet, Googlenet_v2 as well as Resnet_50 for various engines used by the deep learning framework, for various batch sizes. Furthermore,...
-
To Survive in a CBRN Hostile Environment: Application of CAVE Automatic Virtual Environments in First Responder Training
PublicationThis paper is of a conceptual nature and focuses on the use of a specific virtual reality environment in civil-military training. We analyzed the didactic potential of so-called CAVE automatic virtual environments for First Responder training, a type of training that fills the gap between First Aid training and the training received by emergency medical technicians. Since real training involves live drills based on unexpected situations,...
-
A system for singing training
PublicationThe system proposed is aimed at the vocal students and persons who want to improve emission of their voices. The goal is not to substituite a singing teacher but to provide a tool for automatic teaching of voice emission basics. In this way singers can develop their vocal skills and improve them. By a visual feedback a student can control and modify vocal tract maximas (resonances) of a chosen vowel to match the resonances of the...
-
Semantic segmentation training using imperfect annotations and loss masking
PublicationOne of the most significant factors affecting supervised neural network training is the precision of the annotations. Also, in a case of expert group, the problem of inconsistent data annotations is an integral part of real-world supervised learning processes, well-known to researchers. One practical example is a weak ground truth delineation for medical image segmentation. In this paper, we have developed a new method of accurate...
-
Collision‐Aware Routing Using Multi‐Objective Seagull Optimization Algorithm for WSN‐Based IoT
PublicationIn recent trends, wireless sensor networks (WSNs) have become popular because of their cost, simple structure, reliability, and developments in the communication field. The Internet of Things (IoT) refers to the interconnection of everyday objects and sharing of information through the Internet. Congestion in networks leads to transmission delays and packet loss and causes wastage of time and energy on recovery. The routing protocols...
-
Web based training system TeleCAD. Teleworkers training for CAD system users
PublicationW artykule zaprezentowano internetowy system dostarczania kursów do nauczania na odległość. System został opracowany w ramach projektu Leonardo da Vinci TeleCAD (1998-2001). Pokazano jak system wykorzystywany jest obecnie oraz możliwość wykorzystania w projekcie CURE (Research Framework Programme5, 2003-2005).