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 "long term memory" wg kryterium: Temat

  Lista filtrów
Wyrównaj
Wyświetlanie 1-15 z 20
Tytuł:
Adaptive changes in human memory: a literature review
Autorzy:
Sabiniewicz, Agnieszka Laura
Sorokowski, Piotr
Powiązania:
https://bibliotekanauki.pl/articles/954177.pdf
Data publikacji:
2017
Wydawca:
Fundacja Pro Scientia Publica
Tematy:
adaptation
environmental changes
long-term memory
short-term memory
working memory
sensory memory
Opis:
The paper contains a review of the literature concerning memory abilities and human senses performance under different environmental circumstances. A number of studies indicated that environment has a significant impact on human senses functioning. It can affect it in a mechanical way, by a chronic exposure to potentially harmful substances or processes in different work environments. Also, some cognitive abilities that have evolved to perform evolutionary essential functions lost their importance because of the change of environment impact. Moreover, training can be a source of improvement of both human senses and cognitive abilities, as well. That might suggest that, while using, under different environmental circumstances different cognitive abilities develop. We take into a particular consideration human memory and its role, show current studies in this field and suggest new research directions.
Źródło:
Journal of Education Culture and Society; 2017, 8, 1; 79-87
2081-1640
Pojawia się w:
Journal of Education Culture and Society
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
COMPARING THE SHORT AND LONG-TERM EFFECTS OF ACUTE MODERATE-INTENSITY EXERCISE ON MNEMONIC SIMILARITY AND RMOTIONAL MEMORY TASKS
Autorzy:
Acevedo-Triana, Cesar
Cordoba-Patiño, Diana
Muñoz, Juan Francisco
Cifuentes, Julian
Melgarejo Pinto, Victor
Rodriguez W, Oscar
Hurtado-Parrado, Camilo
Powiązania:
https://bibliotekanauki.pl/articles/2137978.pdf
Data publikacji:
2021-01-28
Wydawca:
Fundacja Edukacji Medycznej, Promocji Zdrowia, Sztuki i Kultury Ars Medica
Tematy:
exercise moderate-intensity
pattern separation
emotional memory
long term memory
young
Opis:
Mounting research has linked acute moderate-intensity exercise with changes indiscrimination of similar events – i.e., mnemonic memory. Conversely, few studies have compared performance in tasks associated to each type of memory(mnemonic similarity and emotional) and less have evaluated performance several days after exercise sessions. Thirty-five undergraduate students were randomly distributed in three groups that differed in the assigned duration of the moderate-intensity ex- ercise session. We established first the moderate-intensity exercise program by calculating the VO2max 50%. Two-to-five days later, participants engaged in the exercise condition to which they were assigned, followed by a five-minute rest period. Immediately after, all participants were ex posed to the training phase of both memory tasks. The first retrieval phase was tested 45 minutes after encoding phase was completed. Subsequent retrieval phases were conducted 24, 48, and 168 hours post-training. Exercise of long duration increased discrimination performance in images of low similarity. Comparison of the effects of exercise on discrimination of the three types of images that the emotional-memory task entails showed improved performance only for aversive and neutral images. Exercise improves discrimination of low similarity images, with better overall perform- ance after a longer exercise session. This finding adds to previous reports that have found analogous effects using other memory tasks. It also supports the notion that acute effects due to exercise are specifically related to hippocampal functionality and its ability to separate patterns. Finally, maintenance of emotional informa- tion across time suggest a different mechanism, independent of pattern- separation processing. ----------------------------------------------------------------------------------------------------------------------------
Źródło:
Acta Neuropsychologica; 2021, 19(1); 33-61
1730-7503
2084-4298
Pojawia się w:
Acta Neuropsychologica
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Long-range dependence in DataCenter networks transmission
Autorzy:
Paszkiewicz, A.
Bolanowski, M.
Powiązania:
https://bibliotekanauki.pl/articles/114121.pdf
Data publikacji:
2017
Wydawca:
Stowarzyszenie Inżynierów i Techników Mechaników Polskich
Tematy:
Hurst exponen
long-term memory
network convergence
data center protocols
Opis:
The paper presents the mechanisms of long-range dependence measurement in the context of data transmission in Data Center networks. The research involved mainly analyzing network traffic generated by protocols such as CIFS and iSCSI, which are commonly used in such infrastructures. The purpose of the paper was to determine whether the network traffic of above mentioned protocols encapsulated in TCP/IP protocol will have persistent, anti-persist, or random walk character. By indicating long-range dependencies for this type of network traffic, it will be possible to develop effective mechanisms for detecting anomaly in its transmission as well as flow control, including QoS mechanisms, load balancing, etc.
Źródło:
Measurement Automation Monitoring; 2017, 63, 8; 275-277
2450-2855
Pojawia się w:
Measurement Automation Monitoring
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Concept Maps and Obliteration in Bilinguals
Autorzy:
Gómez Ramos, José Luis
Bravo Palomares, Silvia María
Powiązania:
https://bibliotekanauki.pl/articles/43467352.pdf
Data publikacji:
2024
Wydawca:
Uniwersytet Dolnośląski DSW. Wydawnictwo Naukowe DSW
Tematy:
concept maps
long-term memory
CLIL
bilingual education
instructional design
Opis:
This study examines the effectiveness of concept maps in promoting long-term memory among Content and Language Integrated Learning (CLIL) students. It focuses on the accuracy of content transmission and the acquisition of meaningful learning in bilingual education by connecting new and carefully organized information to students' prior knowledge. Thus, the research assesses the use of concept maps as instructional tools in foreign language (L2, or 'second language') settings, addressing a lack of evidence regarding their effectiveness. It also considers how concept mapping affects long-term memory through factors such as perception, processing, cognition, and transfer. The study examines how bilingualism, bilingual education, and curricular content influence instructional design when using concept maps. The study involved 60 Spanish primary education students attending a semi-public bilingual school. The research results aim to contribute to the development of effective teaching strategies and instructional design in CLIL classrooms, ultimately enhancing students' long-term memory and learning outcomes.
Źródło:
Teraźniejszość – Człowiek – Edukacja; 2024, 26, 1(95); 9-32
1505-8808
2450-3428
Pojawia się w:
Teraźniejszość – Człowiek – Edukacja
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Investment risk assessment based on the long-term memory parameter
Autorzy:
Zeug-Żebro, Katarzyna
Powiązania:
https://bibliotekanauki.pl/articles/1878517.pdf
Data publikacji:
2020
Wydawca:
Politechnika Śląska. Wydawnictwo Politechniki Śląskiej
Tematy:
long-term memory
financial time series
investment risk
local Whittle estimator
pamięć długotrwała
finansowe szeregi czasowe
ryzyko inwestycyjne
Opis:
Purpose: The presence of a long-term memory component in a time series means that even very distant observations exert a certain influence on subsequent implementations of the process. Generally, this relationship is not particularly strong, but it does exist. Interpreting this phenomenon in the context of financial time series, one can come to the conclusion that information that has affected the market some time ago may still be important for the current quotation. The article is devoted to checking the existence of a long-term memory in the financial time series and assessing the investment risk of these series based on the long-term memory parameter. Design/methodology/approach: In order to study the phenomenon of long-term memory in financial time series, the local Whittle estimator was used, while the investment risk assessment was carried out using the fractal dimension, β-coefficient and standard deviation of rates of return. Findings: In the first part of the study the author indicated time series which were characterized by the phenomenon of long-term memory. Then, on the basis of selected measures, the risk of investment was estimated and shares with the least risk were indicated. Research limitations/implications: The results obtained for selected measures showed discrepancies between the shares with the highest and the lowest level of investment risk. Although the results obtained do not give a definite answer which risk measure is more effective, they encourage the use of other measures related to the phenomenon of long-term memory. Practical implications: Application in portfolio analysis. Originality/value: The use of the long-term memory parameter to assess the investment risk of shares.
Źródło:
Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska; 2020, 144; 671-680
1641-3466
Pojawia się w:
Zeszyty Naukowe. Organizacja i Zarządzanie / Politechnika Śląska
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Interval Prediction of Remaining Useful Life based on Convolutional Auto-Encode and Lower Upper Bound Estimation
Autorzy:
Lyu, Yi
Zhang, Qichen
Chen, Aiguo
Wen, Zhenfei
Powiązania:
https://bibliotekanauki.pl/articles/24200839.pdf
Data publikacji:
2023
Wydawca:
Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
Tematy:
remaining useful life
lower upper bound estimation
Long Short-Term Memory
prediction interval
Opis:
Deep learning is widely used in remaining useful life (RUL) prediction because it does not require prior knowledge and has strong nonlinear fitting ability. However, most of the existing prediction methods are point prediction. In practical engineering applications, confidence interval of RUL prediction is more important for maintenance strategies. This paper proposes an interval prediction model based on Long Short-Term Memory (LSTM) and lower upper bound estimation (LUBE) for RUL prediction. First, convolutional auto-encode network is used to encode the multi-dimensional sensor data into one-dimensional features, which can well represent the main degradation trend. Then, the features are input into the prediction framework composed of LSTM and LUBE for RUL interval prediction, which effectively solves the defect that the traditional LUBE network cannot analyze the internal time dependence of time series. In the experiment section, a case study is conducted using the turbofan engine data set CMAPSS, and the advantage is validated by carrying out a comparison with other methods.
Źródło:
Eksploatacja i Niezawodność; 2023, 25, 2; art. no. 165811
1507-2711
Pojawia się w:
Eksploatacja i Niezawodność
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
A novel method of health indicator construction and remaining useful life prediction based on deep learning
Autorzy:
Zhan, Xianbiao
Liu, Zixuan
Yan, Hao
Wu, Zhenghao
Guo, Chiming
Jia, Xisheng
Powiązania:
https://bibliotekanauki.pl/articles/27312791.pdf
Data publikacji:
2023
Wydawca:
Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
Tematy:
stacked sparse autoencoder
health indicator
long short-term memory network
remaining useful life prediction
Opis:
The construction of health indicators (HI) for traditional deep learning requires human training labels and poor interpretability. This paper proposes an HI construction method based on Stacked Sparse Autoencoder (SSAE) and combines SSAE with Long short-term memory (LSTM) network to predict the remaining useful life (RUL). Extracting features from a single domain may result in insufficient feature extraction and cannot comprehensively reflect the degradation status information of mechanical equipment. In order to solve the problem, this article extracts features from time domain, frequency domain, and time-frequency domain to construct a comprehensive original feature set. Based on monotonicity, trendiness, and robustness, the most sensitive features from the original feature set are selected and put into the SSAE network to construct HI for state partitioning, and then LSTM is used for RUL prediction. By comparing with the existing methods, it is proved that the prediction effect of the proposed method in this paper is satisfied.
Źródło:
Eksploatacja i Niezawodność; 2023, 25, 4; art. no. 171374
1507-2711
Pojawia się w:
Eksploatacja i Niezawodność
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Coherence Model of Instruction
Autorzy:
Kapoun, Pavel
Powiązania:
https://bibliotekanauki.pl/articles/448436.pdf
Data publikacji:
2017-04-11
Wydawca:
Wydawnictwo Uniwersytetu Śląskiego
Tematy:
retention of the curriculum in long-term memory
excursion
coherence model of instruction
cooperative learning
mobile learning
museum pedagogy and didactics
understanding in context
spatial learning strategies
Opis:
The article deals with three main issues: the understanding of curriculum in context, the ability of contextualisation, and retention of knowledge in long-term memory. The paper first suggests principles based on the coherence model of instruction, which aims to achieve coherence of knowledge of isolated facts through a network of semantic relationships. Then, the theoretical basis of the model is described, including spatial learning strategies, cooperative learning, and excursions in an authentic environment supported by mobile devices. A methodology of teaching was designed according to the principles of the coherence model, and a virtual guide through educational exhibitions was developed. The virtual guide was tested with students of a primary school during an experimental lecture in the Ostrava Zoo. An evaluation of the coherence model and the virtual guide was carried out using three methods: an observation of students’ behaviour and learning during the experimental lecture, a pedagogical experiment, and an evaluation of questionnaires. The results of the evaluation proved that the coherence model of instruction has a positive impact on understanding in context, ability of contextualisation, and retention of the curriculum in long-term memory.
Źródło:
International Journal of Research in E-learning IJREL; 2016, 2, 2; 81-91
2451-2583
2543-6155
Pojawia się w:
International Journal of Research in E-learning IJREL
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Deep Belief Neural Networks and Bidirectional Long-Short Term Memory Hybrid for Speech Recognition
Autorzy:
Brocki, Ł.
Marasek, K.
Powiązania:
https://bibliotekanauki.pl/articles/177625.pdf
Data publikacji:
2015
Wydawca:
Polska Akademia Nauk. Czytelnia Czasopism PAN
Tematy:
deep belief neural networks
long-short term memory
bidirectional recurrent neural networks
speech recognition
large vocabulary continuous speech recognition
Opis:
This paper describes a Deep Belief Neural Network (DBNN) and Bidirectional Long-Short Term Memory (LSTM) hybrid used as an acoustic model for Speech Recognition. It was demonstrated by many independent researchers that DBNNs exhibit superior performance to other known machine learning frameworks in terms of speech recognition accuracy. Their superiority comes from the fact that these are deep learning networks. However, a trained DBNN is simply a feed-forward network with no internal memory, unlike Recurrent Neural Networks (RNNs) which are Turing complete and do posses internal memory, thus allowing them to make use of longer context. In this paper, an experiment is performed to make a hybrid of a DBNN with an advanced bidirectional RNN used to process its output. Results show that the use of the new DBNN-BLSTM hybrid as the acoustic model for the Large Vocabulary Continuous Speech Recognition (LVCSR) increases word recognition accuracy. However, the new model has many parameters and in some cases it may suffer performance issues in real-time applications.
Źródło:
Archives of Acoustics; 2015, 40, 2; 191-195
0137-5075
Pojawia się w:
Archives of Acoustics
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Detection and Localization of Audio Event for Home Surveillance Using CRNN
Autorzy:
Suruthhi, V. S.
Smita, V.
Rolant Gini, J.
Ramachandran, K. I.
Powiązania:
https://bibliotekanauki.pl/articles/2055274.pdf
Data publikacji:
2021
Wydawca:
Polska Akademia Nauk. Czytelnia Czasopism PAN
Tematy:
convolutional recurrent neural network
CRNN
gated recurrent unit
GRU
long short-term memory
LSTM
sound event localization and detection
SELD
Opis:
Safety and security have been a prime priority in people’s lives, and having a surveillance system at home keeps people and their property more secured. In this paper, an audio surveillance system has been proposed that does both the detection and localization of the audio or sound events. The combined task of detecting and localizing the audio events is known as Sound Event Localization and Detection (SELD). The SELD in this work is executed through Convolutional Recurrent Neural Network (CRNN) architecture. CRNN is a stacked layer of convolutional neural network (CNN), recurrent neural network (RNN) and fully connected neural network (FNN). The CRNN takes multichannel audio as input, extracts features and does the detection and localization of the input audio events in parallel. The SELD results obtained by CRNN with the gated recurrent unit (GRU) and with long short-term memory (LSTM) unit are compared and discussed in this paper. The SELD results of CRNN with LSTM unit gives 75% F1 score and 82.8% frame recall for one overlapping sound. Therefore, the proposed audio surveillance system that uses LSTM unit produces better detection and overall performance for one overlapping sound.
Źródło:
International Journal of Electronics and Telecommunications; 2021, 67, 4; 735--741
2300-1933
Pojawia się w:
International Journal of Electronics and Telecommunications
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Forecasting future values of time series using the lstm network on the example of currencies and WIG20 companies
Prognozowanie przyszłych wartości szeregów czasowych z wykorzystaniem sieci lstm na przykładzie kursów walut i spółek WIG20
Autorzy:
Mróz, Bartosz
Nowicki, Filip
Powiązania:
https://bibliotekanauki.pl/articles/2016294.pdf
Data publikacji:
2020
Wydawca:
Politechnika Bydgoska im. Jana i Jędrzeja Śniadeckich. Wydawnictwo PB
Tematy:
recurrent neural network
RNN
gated recurrent unit
GRU
long short-term memory
LSTM
rekurencyjna sieć neuronowa
blok rekurencyjny
pamięć długookresowa
Opis:
The article presents a comparison of the RNN, GRU and LSTM networks in predicting future values of time series on the example of currencies and listed companies. The stages of creating an application which is a implementation of the analyzed issue were also shown – the selection of networks, technologies, selection of optimal network parameters. Additionally, two conducted experiments were discussed. The first was to predict the next values of WIG20 companies, exchange rates and cryptocurrencies. The second was based on investments in cryptocurrencies guided solely by the predictions of artificial intelligence. This was to check whether the investments guided by the predictions of such a program have a chance of effective earnings. The discussion of the results of the experiment includes an analysis of various interesting phenomena that occurred during its duration and a comprehensive presentation of the relatively high efficiency of the proposed solution, along with all kinds of graphs and comparisons with real data. The difficulties that occurred during the experiments, such as coronavirus or socio-economic events, such as riots in the USA, were also analyzed. Finally, elements were proposed that should be improved or included in future versions of the solution – taking into account world events, market anomalies and the use of supervised learning.
W artykule przedstawiono porównanie sieci RNN, GRU i LSTM w przewidywaniu przyszłych wartości szeregów czasowych na przykładzie walut i spółek giełdowych. Przedstawiono również etapy tworzenia aplikacji będącej realizacją analizowanego zagadnienia – dobór sieci, technologii, dobór optymalnych parametrów sieci. Dodatkowo omówiono dwa przeprowadzone eksperymenty. Pierwszym było przewidywanie kolejnych wartości spółek z WIG20, kursów walut i kryptowalut. Drugi opierał się na inwestycjach w kryptowaluty, kierując się wyłącznie przewidywaniami sztucznej inteligencji. Miało to na celu sprawdzenie, czy inwestowanie na podstawie przewidywania takiego programu pozwala na efektywne zarobki. Omówienie wyników eksperymentu obejmuje analizę różnych ciekawych zjawisk, które wystąpiły w czasie jego trwania oraz kompleksowe przedstawienie relatywnie wysokiej skuteczności proponowanego rozwiązania wraz z wszelkiego rodzaju wykresami i porównaniami z rzeczywistymi danymi. Analizowano również trudności, które wystąpiły podczas eksperymentów, takie jak koronawirus, wydarzenia społeczno-gospodarcze czy zamieszki w USA. Na koniec zaproponowano elementy, które należałoby ulepszyć lub uwzględnić w przyszłych wersjach rozwiązania, uwzględniając wydarzenia na świecie, anomalie rynkowe oraz wykorzystanie uczenia się nadzorowanego.
Źródło:
Zeszyty Naukowe. Telekomunikacja i Elektronika / Uniwersytet Technologiczno-Przyrodniczy w Bydgoszczy; 2020, 24; 13-30
1899-0088
Pojawia się w:
Zeszyty Naukowe. Telekomunikacja i Elektronika / Uniwersytet Technologiczno-Przyrodniczy w Bydgoszczy
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Tool wear condition monitoring in milling process based on data fusion enhanced long short-term memory network under different cutting conditions
Autorzy:
Zheng, Guoxiao
Sun, Weifang
Zhang, Hao
Zhou, Yuqing
Gao, Chen
Powiązania:
https://bibliotekanauki.pl/articles/2038054.pdf
Data publikacji:
2021
Wydawca:
Polska Akademia Nauk. Polskie Naukowo-Techniczne Towarzystwo Eksploatacyjne PAN
Tematy:
tool wear condition monitoring
empirical mode decomposition
variational mode decomposition
fourier synchro squeezed transform
neighborhood component analysis
long short-term memory network
Opis:
Tool wear condition monitoring (TCM) is essential for milling process to ensure the machining quality, and the long short-term memory network (LSTM) is a good choice for predicting tool wear value. However, the robustness of LSTM- based method is poor when cutting condition changes. A novel method based on data fusion enhanced LSTM is proposed to estimate tool wear value under different cutting conditions. Firstly, vibration time series signal collected from milling process are transformed to feature space through empirical mode decomposition, variational mode decomposition and fourier synchro squeezed transform. And then few feature series are selected by neighborhood component analysis to reduce dimension of the signal features. Finally, these selected feature series are input to train the bidirectional LSTM network and estimate tool wear value. Applications of the proposed method to milling TCM experiments demonstrate it outperforms significantly SVR- based and RNN- based methods under different cutting conditions.
Źródło:
Eksploatacja i Niezawodność; 2021, 23, 4; 612-618
1507-2711
Pojawia się w:
Eksploatacja i Niezawodność
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Missing Precipitation Data Estimation Using Long Short-Term Memory Deep Neural Networks
Autorzy:
Djerbouai, Salim
Powiązania:
https://bibliotekanauki.pl/articles/2086428.pdf
Data publikacji:
2022
Wydawca:
Polskie Towarzystwo Inżynierii Ekologicznej
Tematy:
Hodna
K'sob basin
missing precipitation data
long short-term memory
CCWM
coefficient of correlation weighting method
IDWM
inverse distance weighting method
Opis:
Due to the spatiotemporal variability of precipitation and the complexity of physical processes involved, missing precipitation data estimation remains as a significant problem. Algeria, like other countries in the world, is affected by this problem. In the present paper, Long Short-Term Memory (LSTM) deep neural Networks model was tested to estimate missing monthly precipitation data. The application was presented for the K'sob basin, Algeria. In the present paper, the optimal architecture of LSTM model was adjusted by trial-and-error-procedure. The LSTM model was compared with the most widely used classical methods including inverse distance weighting method (IDWM) and the coefficient of correlation weighting method (CCWM). Finally, it was concluded that the LSTM model performed better than the other methods.
Źródło:
Journal of Ecological Engineering; 2022, 23, 5; 216--225
2299-8993
Pojawia się w:
Journal of Ecological Engineering
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
Analysis and Forecasting of the Primary Energy Consumption in Poland Using Deep Learning
Analiza i prognozowanie zużycia energii pierwotnej w Polsce z wykorzystaniem technik głębokiego uczenia
Autorzy:
Manowska, Anna
Powiązania:
https://bibliotekanauki.pl/articles/318083.pdf
Data publikacji:
2020
Wydawca:
Polskie Towarzystwo Przeróbki Kopalin
Tematy:
primary energy consumption
deep learning methods
long short-term memory
deep neural network
konsumpcja energii pierwotnej
metody głębokiego uczenia
sztuczne sieci neuronowe
LSTM
Opis:
Consumption of fossil energy resources were increased dramatically, due to the economic and population growth. In turn, the consumption of fossil resources causes depletion of resources and contributes to environmental pollution. The European Union's "climate neutrality" initiative requires effective energy management from the member states. By this is meant a resource-efficient and competitive economy in which there is no greenhouse gas emission and where economic growth is decoupled from resource consumption. The article analyzes the level of primary energy consumption in Poland. It was examined whether a 23% drop in energy consumption could be achieved in 2030 compared to the base year and according with energy efficiency assumptions. A methodology for forecasting primary energy consumption based on deep neural networks, in particular on Long Short Term Memory (LSTM) algorithms was also presented.
Zużycie kopalnych surowców energetycznych wzrasta, a wzrost ten jest skorelowany ze wzrostem ludności i rozwojem gospodarczym. Z kolei zużycie kopalnych surowców energetycznych powoduje wyczerpywanie się zasobów i przyczynia się do zanieczyszczenia środowiska. Inicjatywa Unii Europejskiej "neutralność klimatyczna" wymaga od państw członkowskich efektywnego zarządzania energią. Przez co rozumie się zasobooszczędną i konkurencyjną gospodarką, w której nie ma emisji netto gazów cieplarnianych i gdzie wzrost gospodarczy jest oddzielony od zużycia zasobów. W artykule przeanalizowano poziom zużycia energii pierwotnej w Polsce. Zbadano, czy w roku 2030 uda się osiągnąć 23% spadek konsumpcji energii w odniesieniu do roku bazowego, zgodnie z przyjętymi założeniami o efektywności energetycznej. Przedstawiono również metodologię prognozowania zużycia energii pierwotnej opartą na głębokich sieciach neuronowych, w szczególności na algorytmach Long Short Term Memory (LSTM).
Źródło:
Inżynieria Mineralna; 2020, 1, 1; 217-222
1640-4920
Pojawia się w:
Inżynieria Mineralna
Dostawca treści:
Biblioteka Nauki
Artykuł
Tytuł:
An optimized parallel implementation of non-iteratively trained recurrent neural networks
Autorzy:
El Zini, Julia
Rizk, Yara
Awad, Mariette
Powiązania:
https://bibliotekanauki.pl/articles/2031147.pdf
Data publikacji:
2021
Wydawca:
Społeczna Akademia Nauk w Łodzi. Polskie Towarzystwo Sieci Neuronowych
Tematy:
GPU implementation
parallelization
Recurrent Neural Network
RNN
Long-short Term Memory
LSTM
Gated Recurrent Unit
GRU
Extreme Learning Machines
ELM
non-iterative training
Opis:
Recurrent neural networks (RNN) have been successfully applied to various sequential decision-making tasks, natural language processing applications, and time-series predictions. Such networks are usually trained through back-propagation through time (BPTT) which is prohibitively expensive, especially when the length of the time dependencies and the number of hidden neurons increase. To reduce the training time, extreme learning machines (ELMs) have been recently applied to RNN training, reaching a 99% speedup on some applications. Due to its non-iterative nature, ELM training, when parallelized, has the potential to reach higher speedups than BPTT. In this work, we present Opt-PR-ELM, an optimized parallel RNN training algorithm based on ELM that takes advantage of the GPU shared memory and of parallel QR factorization algorithms to efficiently reach optimal solutions. The theoretical analysis of the proposed algorithm is presented on six RNN architectures, including LSTM and GRU, and its performance is empirically tested on ten time-series prediction applications. Opt- PR-ELM is shown to reach up to 461 times speedup over its sequential counterpart and to require up to 20x less time to train than parallel BPTT. Such high speedups over new generation CPUs are extremely crucial in real-time applications and IoT environments.
Źródło:
Journal of Artificial Intelligence and Soft Computing Research; 2021, 11, 1; 33-50
2083-2567
2449-6499
Pojawia się w:
Journal of Artificial Intelligence and Soft Computing Research
Dostawca treści:
Biblioteka Nauki
Artykuł
Wyświetlanie 1-15 z 20

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