Informacja

Drogi użytkowniku, aplikacja do prawidłowego działania wymaga obsługi JavaScript. Proszę włącz obsługę JavaScript w Twojej przeglądarce.

English (EN)·Polski (PL)·Yкраїнський (UK)
Przejdź do strony domowej biblioteki Centralna Biblioteka Wojskowa Link otwiera się w nowym oknie Logo biblioteki Prolib Integro - strona głównaCentralna Biblioteka Wojskowa

Wyszukujesz frazę "feed-forward neural network" wg kryterium: Wszystkie pola

  Lista filtrów
Wyrównaj
    Wyświetlanie 1-9 z 9
    Tytuł:
    Feed Forward Neural Network for Autofluorescence Imaging Classification
    Autorzy:
    Kulas, Z.
    Bereś-Pawlik, E.
    Wierzbicki, J.
    Powiązania:
    https://bibliotekanauki.pl/articles/1506801.pdf
    Data publikacji:
    2010-12
    Wydawca:
    Polska Akademia Nauk. Instytut Fizyki PAN
    Tematy:
    87.57.-s
    87.57.R-
    87.57.nm
    87.19.xu
    87.19.xj
    Opis:
    The key elements in cancer diagnostics are the early identification and estimation of the tumor growth and its spread in order to determine the area to be operated on. The aim of our study was to develop new methods of analyzing autofluorescence images which will allow us an objective and accurate assessment of the location of a tumor and will also be helpful in determining the advancement of the disease. The proposed classification methods are based on neural network algorithms. An Olympus company endoscopic system was used for an autofluorescence intestine imaging study. The autofluorescence imaging analysis process can be divided into several main stages. The first step is preparation of a training data set. The second one involves selection of feature space, namely the selection of those features which enable distinguishing the pathologically altered areas from the healthy ones. Final stages of the analysis include pathologically changed tissue classification and diagnosis.
    Źródło:
    Acta Physica Polonica A; 2010, 118, 6; 1189-1193
    0587-4246
    1898-794X
    Pojawia się w:
    Acta Physica Polonica A
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Control of heating processes in electro transport mechatronic system using sigmoidal feed forward neural network
    Autorzy:
    Beinarts, I.
    Levchenkov, A.
    Balckars, P.
    Powiązania:
    https://bibliotekanauki.pl/articles/386280.pdf
    Data publikacji:
    2008
    Wydawca:
    Politechnika Białostocka. Oficyna Wydawnicza Politechniki Białostockiej
    Tematy:
    sieci neuronowe
    transport publiczny
    proces nagrzewania
    neural networks
    public transport
    heating process
    Opis:
    In this article interest is concentrated on the climate parameters optimization in passengers’ interior of mechatronic systems (public electric transportation vehicles- train, tram or trolleybus). Idea is to use feed forward artificial neural network to create an algorithm and coordination mechanism for heating system parameters control to save electrical energy, and to increase the level of comfort for passengers. A special interest for investigations and further development is devoted to intelligent HVAC system allowing more flexible control of the system’s compressor, fan and heater operation, and, therefore, improvement of efficiency and energy saving. This paper provides the mathematical model and algorithm for optimal control of the climate control system.
    Źródło:
    Acta Mechanica et Automatica; 2008, 2, 3; 14-18
    1898-4088
    2300-5319
    Pojawia się w:
    Acta Mechanica et Automatica
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Cascade Feed Forward Neural Network-based Model for Air Pollutants Evaluation of Single Monitoring Stations in Urban Areas
    Autorzy:
    Capizzi, G.
    Lo Sciuto, G.
    Monforte, P.
    Napoli, C.
    Powiązania:
    https://bibliotekanauki.pl/articles/226736.pdf
    Data publikacji:
    2015
    Wydawca:
    Polska Akademia Nauk. Czytelnia Czasopism PAN
    Tematy:
    neural networks
    Synthetic Aperture Radar (SAR)
    mahalanobis distance
    Opis:
    In this paper, air pollutants concentrations for NO2, NO, NOx and PM10 in a single monitoring station are predicted using the data coming from other different monitoring stations located nearby. A cascade feed forward neural network based modeling is proposed. The main aim is to provide a methodology leading to the introduction of virtual monitoring station points consistent with the actual stations located in the city of Catania in Italy.
    Źródło:
    International Journal of Electronics and Telecommunications; 2015, 61, 4; 327-332
    2300-1933
    Pojawia się w:
    International Journal of Electronics and Telecommunications
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Przykład optymalizacji struktury sztucznej sieci neuronowej metodą algorytmów genetycznych
    An example of feed forward neural network structure optimisation with genetic algorithm
    Autorzy:
    Grad, L.
    Powiązania:
    https://bibliotekanauki.pl/articles/273401.pdf
    Data publikacji:
    2006
    Wydawca:
    Wojskowa Akademia Techniczna im. Jarosława Dąbrowskiego
    Tematy:
    sieć neuronowa
    algorytmy genetyczne
    optymalizacja
    neural network
    genetic algorithm
    optimisation
    Opis:
    W artykule przedstawiono przykład optymalizacji struktury jednokierunkowej wielowarstwowej sztucznej sieci neuronowej metodą algorytmów genetycznych. Zaproponowano funkcję przystosowania pozwalającą ocenić jakość proponowanej struktury. Obliczenia wykonano dla sieci neuronowej rozpoznającej cyfry pisane odręcznie.
    An example of feed forward neural network structure optimisation with genetic algorithm is presented. In genetic algorithm an original fitness function is applied. All calculations have been realized for a feed forward neural network, which recognizes hand-written signs.
    Źródło:
    Biuletyn Instytutu Automatyki i Robotyki; 2006, R. 12, nr 23, 23; 27-36
    1427-3578
    Pojawia się w:
    Biuletyn Instytutu Automatyki i Robotyki
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Fast computational approach to the Levenberg-Marquardt algorithm for training feedforward neural networks
    Autorzy:
    Bilski, Jarosław
    Smoląg, Jacek
    Kowalczyk, Bartosz
    Grzanek, Konrad
    Izonin, Ivan
    Powiązania:
    https://bibliotekanauki.pl/articles/2201329.pdf
    Data publikacji:
    2023
    Wydawca:
    Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
    Tematy:
    feed-forward neural network
    neural network learning algorithm
    Levenberg-Marquardt algorithm
    QR decomposition
    Givens rotation
    Opis:
    This paper presents a parallel approach to the Levenberg-Marquardt algorithm (LM). The use of the Levenberg-Marquardt algorithm to train neural networks is associated with significant computational complexity, and thus computation time. As a result, when the neural network has a big number of weights, the algorithm becomes practically ineffective. This article presents a new parallel approach to the computations in Levenberg-Marquardt neural network learning algorithm. The proposed solution is based on vector instructions to effectively reduce the high computational time of this algorithm. The new approach was tested on several examples involving the problems of classification and function approximation, and next it was compared with a classical computational method. The article presents in detail the idea of parallel neural network computations and shows the obtained acceleration for different problems.
    Źródło:
    Journal of Artificial Intelligence and Soft Computing Research; 2023, 13, 2; 45--61
    2083-2567
    2449-6499
    Pojawia się w:
    Journal of Artificial Intelligence and Soft Computing Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Local Levenberg-Marquardt algorithm for learning feedforwad neural networks
    Autorzy:
    Bilski, Jarosław
    Kowalczyk, Bartosz
    Marchlewska, Alina
    Zurada, Jacek M.
    Powiązania:
    https://bibliotekanauki.pl/articles/1837415.pdf
    Data publikacji:
    2020
    Wydawca:
    Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
    Tematy:
    feed-forward neural network
    neural network learning algorithm
    optimization problem
    Levenberg-Marquardt algorithm
    QR decomposition
    Givens rotation
    Opis:
    This paper presents a local modification of the Levenberg-Marquardt algorithm (LM). First, the mathematical basics of the classic LM method are shown. The classic LM algorithm is very efficient for learning small neural networks. For bigger neural networks, whose computational complexity grows significantly, it makes this method practically inefficient. In order to overcome this limitation, local modification of the LM is introduced in this paper. The main goal of this paper is to develop a more complexity efficient modification of the LM method by using a local computation. The introduced modification has been tested on the following benchmarks: the function approximation and classification problems. The obtained results have been compared to the classic LM method performance. The paper shows that the local modification of the LM method significantly improves the algorithm’s performance for bigger networks. Several possible proposals for future works are suggested.
    Źródło:
    Journal of Artificial Intelligence and Soft Computing Research; 2020, 10, 4; 299-316
    2083-2567
    2449-6499
    Pojawia się w:
    Journal of Artificial Intelligence and Soft Computing Research
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Efficient dead time correction of G-M counters using feed forward artificial neural network
    Autorzy:
    Arkani, M.
    Khalafi, A.
    Powiązania:
    https://bibliotekanauki.pl/articles/146121.pdf
    Data publikacji:
    2013
    Wydawca:
    Instytut Chemii i Techniki Jądrowej
    Tematy:
    dead time
    artificial neural network (ANN)
    Geiger-Müller (G-M) detector
    hybrid model
    source decaying experiment
    Opis:
    Dead time parameter of Geiger-Müller (G-M) counters causes a great uncertainty in their response to the incident radiation intensity at high counting rates. As their applications in experimental nuclear science are widespread, many attempts have been done on improvements of their nonlinear response. In this work, response of a G-M counter system is optimized and corrected efficiently using feed forward artificial neural network (ANN). This method is simple, fast, and provides the answer to the problem explicitly with no need for iteration. The method is applied to a set of decaying source experimental data measured by a fairly large G-M tube. The results are compared with those predicted by a given analytical model which is called hybrid model. The maximum deviation of the corrected results from the true counting rates is less than 4% which is a significant improvement in comparison with the results obtained by the analytical method. Results of this study show that by using a proper artificial neural network structure, the dead time effects of G-M counters can be tolerated significantly.
    Źródło:
    Nukleonika; 2013, 58, 2; 317-321
    0029-5922
    1508-5791
    Pojawia się w:
    Nukleonika
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Porównanie podejścia aproksymującego i klasyfikującego w prognozowaniu kursów wybranych akcji na GPW w Warszawie S.A. z użyciem jednokierunkowych sieci neuronowych
    Forecasting Stock Prices Using Feed-Forward Neural Network - a Comparison of Approximation and Classification Approaches
    Autorzy:
    Kasznia, Anna
    Powiązania:
    https://bibliotekanauki.pl/articles/589117.pdf
    Data publikacji:
    2013
    Wydawca:
    Uniwersytet Ekonomiczny w Katowicach
    Tematy:
    Giełda papierów wartościowych
    Kurs akcji
    Prognozowanie
    Sieci neuronowe
    Szeregi czasowe
    Forecasting
    Neural networks
    Share price
    Stock market
    Time-series
    Opis:
    In this paper two approaches to financial time series forecasting using neural networks were compared. First one, the function approximation approach, in which neural networks are trained to forecast the exact one day ahead value of stock price. And the second one, classification approach, in which the output variable is the direction of future stock price movements. The aim of this work was to check if using the classification models can lead to better results in terms of direction of change forecasting and profits generated by their forecasts. This research was conducted on the basis of the time series of daily closing stock prices for three companies listed on the Warsaw Stock Exchange. Simulations show that some of the approximating models achieved satisfactory results in terms of the directional symmetry measure, although the best results for each of the analyzed company have been achieved for classification models.
    Źródło:
    Studia Ekonomiczne; 2013, 146; 59-67
    2083-8611
    Pojawia się w:
    Studia Ekonomiczne
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
    Tytuł:
    Application of Box-Jenkins method and Artificial Neural Network procedure for time series forecasting of prices
    Autorzy:
    Singh, Abhishek
    Mishra, G. C.
    Powiązania:
    https://bibliotekanauki.pl/articles/465876.pdf
    Data publikacji:
    2015
    Wydawca:
    Główny Urząd Statystyczny
    Tematy:
    forecasting
    feed forward network
    ARIMA
    ANN
    Opis:
    Forecasting of prices of commodities, especially those of agricultural commodities, is very difficult because they are not only governed by demand and supply but also by so many other factors which are beyond control, such as weather vagaries, storage capacity, transportation, etc. In this paper time series models namely ARIMA (Autoregressive Integrated Moving Average) methodology given by Box and Jenkins has been used for forecasting prices of Groundnut oil in Mumbai. This approach has been compared with ANN (Artificial Neural Network) methodology. The results showed that ANN performed better than the ARIMA models in forecasting the prices.
    Źródło:
    Statistics in Transition new series; 2015, 16, 1; 83-96
    1234-7655
    Pojawia się w:
    Statistics in Transition new series
    Dostawca treści:
    Biblioteka Nauki
    Artykuł
      Wyświetlanie 1-9 z 9

      Ta witryna wykorzystuje pliki cookies do przechowywania informacji na Twoim komputerze. Pliki cookies stosujemy w celu świadczenia usług na najwyższym poziomie, w tym w sposób dostosowany do indywidualnych potrzeb. Korzystanie z witryny bez zmiany ustawień dotyczących cookies oznacza, że będą one zamieszczane w Twoim komputerze. W każdym momencie możesz dokonać zmiany ustawień dotyczących cookies