Search
Filters
total: 1460
displaying 1000 best results Help
Search results for: multi-gpu scalability
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublicationRecent years have seen impressive progress in visual recognition on many benchmarks, however, generalization to the out-of-distribution setting remains a significant challenge. A state-of-the-art method for robust visual recognition is model ensembling. However, recently it was shown that similarly competitive results could be achieved with a much smaller cost, by using multi-input multi-output architecture (MIMO). In this work,...
-
Robust Object Detection with Multi-input Multi-output Faster R-CNN
PublicationRecent years have seen impressive progress in visual recognition on many benchmarks, however, generalization to the out-of-distribution setting remains a significant challenge. A state-of-the-art method for robust visual recognition is model ensembling. However, recently it was shown that similarly competitive results could be achieved with a much smaller cost, by using multi-input multi-output architecture (MIMO). In this work,...
-
Multi Queue Approach for Network Services Implemented for Multi Core CPUs
PublicationMultiple core processors have already became the dominant design for general purpose CPUs. Incarnations of this technology are present in solutions dedicated to such areas like computer graphics, signal processing and also computer networking. Since the key functionality of network core components is fast package servicing, multicore technology, due to multi tasking ability, seems useful to support packet processing. Dedicated...
-
GPU-Accelerated LOBPCG Method with Inexact Null-Space Filtering for Solving Generalized Eigenvalue Problems in Computational Electromagnetics Analysis with Higher-Order FEM
PublicationThis paper presents a GPU-accelerated implementation of the Locally Optimal Block Preconditioned Conjugate Gradient (LOBPCG) method with an inexact nullspace filtering approach to find eigenvalues in electromagnetics analysis with higherorder FEM. The performance of the proposed approach is verified using the Kepler (Tesla K40c) graphics accelerator, and is compared to the performance of the implementation based on functions from...
-
Multi-objective optimization of microextraction procedures
PublicationOptimization of extraction process requiresfinding acceptable conditions for many analytes and goodperformance in terms of process time or solvent consumption. These optimization criteria are oftencontradictory to each other, the performance of the system in given conditions is good for some criteriabut poor for others. Therefore, such problems require special assessment tools that allow to combinethese contradictory criteria into...
-
Vident-lab: a dataset for multi-task video processing of phantom dental scenes
Open Research DataWe introduce a new, asymmetrically annotated dataset of natural teeth in phantom scenes for multi-task video processing: restoration, teeth segmentation, and inter-frame homography estimation. Pairs of frames were acquired with a beam splitter. The dataset constitutes a low-quality frame, its high-quality counterpart, a teeth segmentation mask, and...
-
TensorHive: Management of Exclusive GPU Access for Distributed Machine Learning Workloads
PublicationTensorHive is a tool for organizing work of research and engineering teams that use servers with GPUs for machine learning workloads. In a comprehensive web interface, it supports reservation of GPUs for exclusive usage, hardware monitoring, as well as configuring, executing and queuing distributed computational jobs. Focusing on easy installation and simple configuration, the tool automatically detects the available computing...
-
Runtime Visualization of Application Progress and Monitoring of a GPU-enabled Parallel Environment
PublicationThe paper presents design, implementation and real life uses of a visualization subsystem for a distributed framework for parallelization of workflow-based computations among clusters with nodes that feature both CPUs and GPUs. Firstly, the proposed system presents a graphical view of the infrastructure with clusters, nodes and compute devices along with parameters and runtime graphs of load, memory available, fan speeds etc. Secondly,...
-
Multi-task Video Enhancement for Dental Interventions
PublicationA microcamera firmly attached to a dental handpiece allows dentists to continuously monitor the progress of conservative dental procedures. Video enhancement in video-assisted dental interventions alleviates low-light, noise, blur, and camera handshakes that collectively degrade visual comfort. To this end, we introduce a novel deep network for multi-task video enhancement that enables macro-visualization of dental scenes. In particular,...
-
Multi-DBD actuator with floating interelectrode for aerodynamic control
PublicationIn this paper the use of a floating inter-electrode in a multi-DBD (dielectric barrier discharge) plasma actuator is described. The multi-DBD plasma actuators with floating inter-electrodes were investigated to get a longer DBD on a dielectric surface and to maximise generated net airflow. Our actuator was used to control the boundary layer flow separation around NACA0012 airfoil model. The results of our investigations suggests...
-
Shared multi-processor scheduling
PublicationWe study shared multi-processor scheduling problem where each job can be executed on its private processor and simultaneously on one of many processors shared by all jobs in order to reduce the job’s completion time due to processing time overlap. The total weighted overlap of all jobs is to be maximized. The problem models subcontracting scheduling in supply chains and divisible load scheduling in computing. We show that synchronized...
-
Partially Isolated Multi-Active Bridge DC-DC Converter with Bidirectional EV Charging Ports
PublicationThere is a growing interest in multiport converters due to their ability to interconnect various energy sources and loads using reduced number of components. Regarding electric vehicle (EV) chargers connection, existing multiport solutions either lack scalability, featuring only one isolated charging port or the operation of ports depend on each other, complicating the converter control algorithm. This article proposes a partially...
-
The OptD-multi method in LiDAR processing
PublicationNew and constantly developing technology for acquiring spatial data, such as LiDAR (light detection and ranging), is a source for large volume of data. However, such amount of data is not always needed for developing the most popular LiDAR products: digital terrain model (DTM) or digital surface model. Therefore, in many cases, the number of contained points are reduced in the pre-processing stage. The degree of reduction is determined...
-
Towards an efficient multi-stage Riemann solver for nuclear physics simulations
PublicationRelativistic numerical hydrodynamics is an important tool in high energy nuclear science. However, such simulations are extremely demanding in terms of computing power. This paper focuses on improving the speed of solving the Riemann problem with the MUSTA-FORCE algorithm by employing the CUDA parallel programming model. We also propose a new approach to 3D finite difference algorithms, which employ a GPU that uses surface memory....
-
AUTONOMOUS AGENTS AND MULTI-AGENT SYSTEMS
Journals -
Journal of Multi-Criteria Decision Analysis
Journals -
Multi-objective weather routing of sailing vessels
PublicationThe paper presents a multi-objective deterministic method of weather routing for sailing vessels. Depending on a particular purpose of sailboat weather routing, the presented method makes it possible to customize the criteria and constraints so as to fit a particular user’s needs. Apart from a typical shortest time criterion, safety and comfort can also be taken into account. Additionally, the method supports dynamic weather data:...
-
Machine Learning in Multi-Agent Systems using Associative Arrays
PublicationIn this paper, a new machine learning algorithm for multi-agent systems is introduced. The algorithm is based on associative arrays, thus it becomes less complex and more efficient substitute of artificial neural networks and Bayesian networks, which is confirmed by performance measurements. Implementation of machine learning algorithm in multi-agent system for aided design of selected control systems allowed to improve the performance...
-
Improved conformational space annealing method to treat β-structure with the UNRES force-field and to enhance scalability of parallel implementation
Publication -
Fast multi-objective design optimization of microwave and antenna structures using data-driven surrogates and domain segmentation
PublicationPurpose Strategies and algorithms for expedited design optimization of microwave and antenna structures in multi-objective setup are investigated. Design/methodology/approach Formulation of the multi-objective design problem oriented towards execution of the population-based metaheuristic algorithm within the segmented search space is investigated. Described algorithmic framework exploit variable fidelity modeling, physics- and...
-
EVOLUTIONARY MULTI–OBJECTIVE WEATHER ROUTING OF SAILBOATS
PublicationThe paper presents a multi-objective method, which optimises the route of a sailboat. The presented method makes use of an evolutionary multi-objective (EMO) algorithm, which performs the optimisation according to three objective functions: total passage time, a sum of all course alterations made during the voyage and the average angle of heel. The last two of the objective functions reflect the navigator’s and passenger’s comfort,...
-
Gaze tracking in multi-display environment
PublicationThis paper presents the basic ideas of eye and gaze tracking in multiple-display environment. The algorithm for display detection and identification is described as well as the rules for gaze interaction in multi display environment. The core of the method is to use special LED markers and eye and scene tracking glasses. Scene tracking camera registers markers position which is then represented as a cloud of points. Analyzing the...
-
Multi-objective Weather Routing with Customised Criteria and Constraints
PublicationThe paper presents a weather routing algorithm utilising a multi-objective optimisation with constraints, namely the Multi-objective Evolutionary Weather Routing Algorithm (MEWRA). In the proposed approach weather route recommendations can be made simultaneously e.g. for passage time, fuel consumption and safety of passage by means of Pareto optimisation. The sets of criteria and constraints in the optimisation process are fully...
-
Multi-criterion, evolutionary and quantum decision making in complex systems
PublicationMulti-criterion, evolutionary and quantum decision making supported by the Adaptive Quantum-based Multi-criterion Evolutionary Algorithm (AQMEA) has been considered for distributed complex systems. AQMEA had been developed to the task assignment problem, and then it has been applied to underwater vehicle planning as another benchmark three-criterion optimization problem. For evaluation of a vehicle trajectory three criteria have...
-
Multi-headed chimera states in coupled pendula
PublicationWe discuss the occurrence of the chimera states in the network of coupled, excited by the clock’s mechanisms pendula. We find the patterns of multi-headed chimera states in which pendula clustered in different heads behave differently (oscillate with different frequencies) and create different types of synchronous states (complete or phase synchronization). The mathematical model of the network shows that the observed chimera states...
-
A multi-agent method for periodicity detection in distributed events
PublicationMulti-agent systems working in constantly changing environments may be challenged by instantaneous unavailability of their autonomous agents caused e.g. by limited computing resources. A new method of self-organization of distributed service components is proposed, suitable for multi-agent systems. This method relies on particular agents carrying out separate analyzes of their individual processing loads or other specific events....
-
Daylight evaluation for multi-family housing in Poland
PublicationThis PhD dissertation focuses on methods of daylight appraisal useful in the design of the contemporary multifamily housing. The theoretical part of the thesis offers a review of daylight indicators, evaluations methods and tools within the built environment. It covers a review of daylight recommendations found in building standards and other normative documents affecting the design of the residential spaces. A pilot work survey...
-
MULTI-CRITERIA MODEL IN MULTIFUNCTIONAL BUILDING SYSTEM DESIGN PROCESS
PublicationThe paper presents a multi-criteria approach in multifunctional building system design process. The aim is to develop a theory relative to the engineering system of multifunctional with a mathematical representation defined by a holistic network for the lifecycle of the designed object. The idea of work was to define the structure of a complex system. Background for the presented field is to develop a design strategy for multifunctional...
-
Selected manufacturing difficulties encountered during setup of machining on cnc multi-axis linear automatic lathe and on cnc multi-spindle turning centers
PublicationThe article presents the observed manufacturing technology implementation difficulties in workshop practice, resulting from the construction and principles of operation on the DMG’s SPRINT 32/5 CNC linear automatic lathe and on the Mazak’s HQR 150 MSY and QTN 200 MS CNC turn - mill centers, and discusses possible rules for solving the production problems encountered. The article also discusses the principles of dividing the machining...
-
Implementation of algebraic procedures on the GPU using CUDA architecture on the example of generalized eigenvalue problem
Publication -
Properties of an innovative multi-functional finish for the improvement of indoor air quality
PublicationDue to lifestyle changes, people spend most of their time indoors at present; thus, Indoor Air Quality (IAQ) is a matter of utmost importance. Multi-functional and innovative finishes can help to passively improve the IAQ, benefitting the health and comfort of occupants. For this study, reference and pre-mixed commercial mortars are compared to a new multi-functional hydraulic lime mortar for indoor finishes, in which conventional...
-
Gender approach to multi-objective optimization of detection systems by pre-selection of criteria
PublicationA novel idea of performing evolutionary computations for solving highly-dimensional multi-objective optimization (MOO) problems is proposed. The information about individual genders is applied. This information is drawn out of the fitness of individuals and applied during the parental crossover in the evolutionary multi-objective optimization (EMO) processes. The paper introduces the principles of the genetic-gender approach (GGA)...
-
Gender approach to multi-objective optimization of detection systems by pre-selection of criteria
PublicationA novel idea of performing evolutionary computations for solving highly-dimensional multi-objective optimization (MOO) problems is proposed. The information about individual genders is applied. This information is drawn out of the fitness of individuals and applied during the parental crossover in the evolutionary multi-objective optimization (EMO) processes. The paper introduces the principles of the genetic-gender approach (GGA)...
-
Gender approach to multi-objective optimization of detection systems by pre-selection of criteria
PublicationA novel idea of performing evolutionary computations for solving highly-dimensional multi-objective optimization (MOO) problems is proposed. The information about individual genders is applied. This information is drawn out of the fitness of individuals and applied during the parental crossover in the evolutionary multi-objective optimization (EMO) processes. The paper introduces the principles of the genetic-gender approach (GGA)...
-
Ontology management in the multi-agent system for the it evaluation
PublicationThe team of researchers at the Gdańsk University of Technology is currently developing a model of the multi-agent system (MAS_IT) that would aid the IT managers to find the most appropriate technology to the goals assumed. The knowledge is to be provided by experts and stored in the facts and rules bases. The responsibility of the verification process-on the input and query stages-is given to the ontology, making it to some extend...
-
Systematic Management of Simulation State for Multi-Branch Simulations in Simulink
PublicationSystematic simulation is a technique related and motivated by the formal analysis of hybrid dynamic systems. It combines the exhaustive and conservative nature of traditional model checking with numerical simulation for providing efficient algorithms to manage simulations. Multi-branch simulation is the concept advancing simulation efficiency by reducing the number of state transitions. This paper introduces an approach to implement...
-
Multi-fidelity EM simulations and constrained surrogate modelling for low-cost multi-objective design optimisation of antennas
PublicationIn this study, a technique for low-cost multi-objective design optimisation of antenna structures has been proposed. The proposed approach is an enhancement of a recently reported surrogate-assisted technique exploiting variable-fidelity electromagnetic (EM) simulations and auxiliary kriging interpolation surrogate, the latter utilised to produce the initial approximation of the Pareto set. A bottleneck of the procedure for higher-dimensional...
-
Multi-Transformer Flyback Converter for Supplying Isolated IGBT and MOSFET Drivers
PublicationA multi transformer flyback converter topology for supplying transistor drivers is presented. The topology presents some advantages over typical multi output single transformer, as reduction of effective leakage inductance, equal magnetic coupling between primary and secondary circuits and better isolation between outputs. Simulation study carried out in the LTSpice IV program and preliminary experimental results indicate high...
-
Multi-objective design of miniaturized impedance transformers by domain segmentation
PublicationFast multi-objective design optimization of compact microstrip impedance transformers is discussed. Our approach exploits approximation models constructed using sampled coarse- mesh EM simulation data in a partitioned design space and response correction techniques for design refinement. Demonstra
-
Multi-criterion decision making in distributed systems by quantum evolutionary algorithms
PublicationDecision making by the AQMEA (Adaptive Quantum-based Multi-criterion Evolutionary Algorithm) has been considered for distributed computer systems. AQMEA has been extended by a chromosome representation with the registry of the smallest units of quantum information. Evolutionary computing with Q-bit chromosomes has been proofed to characterize by the enhanced population diversity than other representations, since individuals represent...
-
From Scores to Predictions in Multi-Label Classification: Neural Thresholding Strategies
PublicationIn this paper, we propose a novel approach for obtaining predictions from per-class scores to improve the accuracy of multi-label classification systems. In a multi-label classification task, the expected output is a set of predicted labels per each testing sample. Typically, these predictions are calculated by implicit or explicit thresholding of per-class real-valued scores: classes with scores exceeding a given threshold value...
-
The new investing effectiveness evaluation multi-criteria method in modern supply systems
PublicationThe important problem in the processes of modelling and programming the development of sustainable energy sector is the multi-criteria manner of assessing the effectiveness of investments. The goal of this paper is to show how to take into account the impact of investments in multidimensional modelling decision-making processes. This goal can be achieved through the development, presentation, and use of a new multi-criteria method...
-
Multi-Criteria Approach in Multifunctional Building Design Process
PublicationThe paper presents new approach in multifunctional building design process. Publication defines problems related to the design of complex multifunctional buildings. Currently, contemporary urban areas are characterized by very intensive use of space. Today, buildings are being built bigger and contain more diverse functions to meet the needs of a large number of users in one capacity. The trends show the need for recognition of...
-
Ranking of Generation Source Locations by a Hybrid Multi-Criteria Method
PublicationThe paper presents a ranking of the locations of eight renewable energy sources (RES) made using a hybrid multi-criteria analysis method. The method is a combination of the analytical hierarchical process (AHP) method and numerical taxonomy. The considered generating sources, i.e. solar plants, biogas plants, and wind farms are sources that will significantly contribute to implementing the provisions of the energy and climate package...
-
An application of multi-agent system for ship’s power systems design
PublicationDesign process of transport ship power system consists of structure (topology) and component elements selection. Compliance with the requirements for static components does not guarantee optimal dynamic characteristics of entire power system. Design steps are difficult to formalize and as a consequence to this reason expert and multi-agent systems are used for solving selected design issues. In the paper distributed multi-agent...
-
MELUS: Multi-Ethnic Literature of the United States
Journals -
IEEE Transactions on Multi-Scale Computing Systems
Journals -
Compact broadband multi-way 1:6 power divider
PublicationThis paper introduces a design flow of a microstrip multi-way 1:6 power divider incorporating photonic bandgap (PBG) structures. Themethod proposed has enabled the achievement of considerable miniaturization (15%) together with transmission characteristics enhancement(141% bandwidth). Measured results show significant similarity to theoretical characteristics, which proves the attractiveness of presented design methodology.
-
History-based dynamic weight voting for multi-agent systems
PublicationTaking decision in multi-agent systems may involve voting. Two basic voting protocols are surveyed in the paper. Then, a new history-based dynamic weight voting is proposed. This voting protocol allows for identification of the agents which contribute to the correct system decision. The proposed solution is contrasted with majority voting and with weighted average voting, to show its advantages
-
Solving highly-dimensional multi-objective optimization problems by means of genetic gender
PublicationPaper presents a computational optimization study using a genetic gender approach for solving multi-objective optimization problems of detection observers. In this methodology the information about an individual gender of all the considered solutions is applied for the purpose of making distinction between different groups of objectives. This information is drawn out of the fitness of individuals and applied during a current parental...