Wyniki wyszukiwania - Biblioteka Nauki

1

Robust zeroing neural networks with two novel power-versatile activation functions for solving dynamic Sylvester equation

100%

Zhou P. , Tan M.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2022

|

tom Vol. 70, nr 3

art. no. e141307

EN

In this work, two robust zeroing neural network (RZNN) models are presented for online fast solving of the dynamic Sylvester equation (DSE), by introducing two novel power-versatile activation functions (PVAF), respectively. Differing from most of the zeroing neural network (ZNN) models activated by recently reported activation functions (AF), both of the presented PVAF-based RZNN models can achieve predefined time convergence in noise and disturbance polluted environment. Compared with the exponential and finite-time convergent ZNN models, the most important improvement of the proposed RZNN models is their fixed-time convergence. Their effectiveness and stability are analyzed in theory and demonstrated through numerical and experimental examples.

2

Comparison of optimization algorithms of connectionist temporal classifier for speech recognition system

100%

Amirgaliyev Y. , Darkhan K. , Shoiynbek A.

Informatyka, Automatyka, Pomiary w Gospodarce i Ochronie Środowiska

|

2019

|

tom T. 9, nr 3

54--57

EN

This paper evaluates and compares the performances of three well-known optimization algorithms (Adagrad, Adam, Momentum) for faster training the neural network of CTC algorithm for speech recognition. For CTC algorithms recurrent neural network has been used, specifically Long- Short-Term memory. LSTM is effective and often used model. Data has been downloaded from VCTK corpus of Edinburgh University. The results of optimization algorithms have been evaluated by the Label error rate and CTC loss.

PL

W artykule dokonano oceny i porównania wydajności trzech znanych algorytmów optymalizacyjnych (Adagrad, Adam, Momentum) w celu przyspieszenia treningu sieci neuronowej algorytmu CTC do rozpoznawania mowy. Dla algorytmów CTC wykorzystano rekurencyjną sieć neuronową, w szczególności LSTM, która jest efektywnym i często używanym modelem. Dane zostały pobrane z wydziału VCTK Uniwersytetu w Edynburgu. Wyniki algorytmów optymalizacyjnych zostały ocenione na podstawie wskaźników Label error i CTC loss.

3

A segment-wise reconstruction method based on bidirectional long short term memory for Power Line Interference suppression

100%

Qiu Y. , Huang K. , Xiao F. , Shen H.

Biocybernetics and Biomedical Engineering

|

2018

|

tom Vol. 38, no. 2

217--227

EN

The overlap between the signal components of Power Line Interference (PLI) and biomedical signals in the frequency domain makes the filtered results prone to severe distortion. Electrocardiogram (ECG) is a type of biomedical electronic signal used for cardiac diagnosis. The objective of this work is to suppress the PLI components from biomedical signals with minimal distortion, and the object of study is mainly the ECG signals. In this study, we propose a novel segment-wise reconstruction method to suppress the PLI in biomedical signals based on the Bidirectional Recurrent Neural Networks with Long Short Term Memory (Bi-LSTM). Experiments are conducted on both synthetic and real signals, and quantitative comparisons are made with a traditional IIR notch filter and two state-of-the-art methods in the literature. The results show that by our method, the output Signal-to-Noise Ratio (SNR) is improved by more than 7 dB and the settling time for step response is reduced to 0.09 s on average. The results also demonstrate that our method has enough generalization ability for unforeseen signals without retraining.

4

Nonlinear model predictive control of a boiler unit: a fault tolerant control study

100%

Patan K. , Korbicz J.

International Journal of Applied Mathematics and Computer Science

|

2012

|

tom Vol. 22, no. 1

225-237

EN

This paper deals with a nonlinear model predictive control designed for a boiler unit. The predictive controller is realized by means of a recurrent neural network which acts as a one-step ahead predictor. Then, based on the neural predictor, the control law is derived solving an optimization problem. Fault tolerant properties of the proposed control system are also investigated. A set of eight faulty scenarios is prepared to verify the quality of the fault tolerant control. Based of different faulty situations, a fault compensation problem is also investigated. As the automatic control system can hide faults from being observed, the control system is equipped with a fault detection block. The fault detection module designed using the one-step ahead predictor and constant thresholds informs the user about any abnormal behaviour of the system even in the cases when faults are quickly and reliably compensated by the predictive controller.

5

A novel deep LSTM network for artifacts detection in microelectrode recordings

100%

Hosny M. , Zhu M. , Gao W. , Fu Y.

Biocybernetics and Biomedical Engineering

|

2020

|

tom Vol. 40, no. 3

1052--1063

EN

Microelectrode recording (MER) signals are world-widely used for validating the planned trajectories in the procedure of deep brain stimulation (DBS) surgery to obtain accurate position of electrodes inside the brain structure. Besides, MER signals are important source for studying extracellular neuronal activity and DBS biomarkers, such as, spike clustering and sorting. However, MER signals are prone to several artifacts derived from electrical equipment in the operating room, electrode movement and patient activities, etc., which reduce the signal-to-noise ratio of the MER signals. Therefore, in this paper, we propose a novel deep learning architecture based on long short-term memory (LSTM) network for automatic artifact detection in MER signals. Frequency and time-domain features were extracted from the raw MER signals and fed to the deep LSTM network. A manually annotated MER database obtained from 17 Parkinson's disease (PD) patients were used to validate the proposed architecture. The proposed architecture achieved promising results of 97.49% accuracy, 98.21% sensitivity and 96.87% specificity on an unseen test set. To our best knowledge, this is the first study to use LSTM network for artifacts detection in MER signals. The MER data will be available at http://homepage.hit.edu.cn/wpgao.

6

Inter-patient ECG classification with convolutional and recurrent neural networks

100%

Guo L. , Sim G. , Matuszewski B.

Biocybernetics and Biomedical Engineering

|

2019

|

tom Vol. 39, no. 3

868--879

EN

The recent advances in ECG sensor devices provide opportunities for user self-managed auto-diagnosis and monitoring services over the internet. This imposes the requirements for generic ECG classification methods that are inter-patient and device independent. In this paper, we present our work on using the densely connected convolutional neural network (DenseNet) and gated recurrent unit network (GRU) for addressing the inter-patient ECG classification problem. A deep learning model architecture is proposed and is evaluated using the MIT-BIH Arrhythmia and Supraventricular Databases. The results obtained show that without applying any complicated data pre-processing or feature engineering methods, both of our models have considerably outperformed the state-of-the-art performance for supraventricular (SVEB) and ventricular (VEB) arrhythmia classifications on the unseen testing dataset (with the F1 score improved from 51.08 to 61.25 for SVEB detection and from 88.59 to 89.75 for VEB detection respectively). As no patient-specific or device-specific information is used at the training stage in this work, it can be considered as a more generic approach for dealing with scenarios in which varieties of ECG signals are collected from different patients using different types of sensor devices.

7

Zastosowanie sterowania z modelem wewnętrznym opartego o sieci neuronowe dla napędu z elastycznym sprzęgłem

100%

Żychlewicz M. , Derugo P.

Przegląd Elektrotechniczny

|

2018

|

tom R. 94, nr 5

163--168

PL

W artykule opisano sterowanie układem napędowym z połączeniem sprężystym, pętla regulacji prędkości została zaprojektowana w oparciu o dwa modele neuronowe. Jeden z nich stanowi główny regulator, natomiast drugi jest modelem odniesienia wykorzystywanym w trakcie obliczeń. Adaptacja wag sieci neuronowych jest realizowana on-line. Artykuł zawiera opis teoretyczny zaimplementowanej struktury, a także badania symulacyjne oraz eksperymentalne zrealizowane z wykorzystaniem procesora sygnałowego karty dSPACE1103.

EN

Paper presents control system applied for electrical drive with elastic connections. Speed control loop of the whole structure is based on two neural models. One of them is applied as the main controller, the second is the internal model of the plant used for calculations of control signal. Adaptation of weights in neural networks is done in on-line mode. Article contains theoretical description of implemented control structure, simulation tests as well as experimental tests using digital signal processor of dSPACE1103.

8

Predykacja zajętości kanału LTE za pomocą rekurencyjnej sieci neuronowej

80%

Wasilewska M. , Bogucka H.

|

2020

|

tom nr 7-8

269--273, CD

PL

W artykule przedstawiono metody detekcji i przewidywania obecności transmisji sygnału LTE. Zastosowano algorytmy uczenia maszynowego, takie jak algorytm k najbliższych sąsiadów, drzewo decyzyjne, sieć neuronową oraz rekurencyjną sieć neuronową. Za pomocą eksperymentów wykazano, że wymienione algorytmy, a w szczególności sieć rekurencyjna osiągają wysokie wartości prawdopodobieństwa poprawnej detekcji zarówno dla detekcji w czasie rzeczywistym, jak i dla przewidywania obecności sygnałów w przyszłości.

EN

In the paper, the methods of LTE spectrum detection and future state predictions have been presented. Machine learning algorithms have been implemented for spectrum sensing, namely k-nearest neighbors, decision tree, neural network and recurrent neural network. Conducted experiment has shown that these algorithms reach high values of probability of correct detection for current moment as well as for future prediction.

9

Human activity recognition using improved complete ensemble EMD with adaptive noise and long short-term memory neural networks

80%

Altuve M. , Lizarazo P. , Villamizar J.

Biocybernetics and Biomedical Engineering

|

2020

|

tom Vol. 40, no. 3

901--909

EN

The recognition of human activities is a topic of great relevance due to its wide range of applications. Different approaches have been proposed to recognize human activities, ranging from the comparison of signals with thresholds to the application of deep and machine learning techniques. In this work, the classification of six human activities (walking, walking downstairs, walking upstairs, standing, sitting, and lying down) is performed using bidirectional LSTM networks that exploit intrinsic mode function (IMF) representation of inertial signals. Records with inertial signals (accelerometer and gyroscope) of 2.56 s, available at the UCI Machine Learning Repository, were collected from 30 subjects using a smartphone. First, inertial signals were standardized to take them to the same scale and were decomposed into IMF using the improved complete ensemble empirical mode decomposition with adaptive noise (ICEEMDAN). IMF were then segmented (split) into nine segments of 1.28 s with 12.5% overlap and introduced to a first network with four outputs to identify the dynamic activities and the statics as a single class called ‘‘statics’’, giving 98.86% accuracy. Then, the non-segmented IMF of the records assigned to the statics class were introduced to a second network to classify their three activities, giving an accuracy of 88.46%. In total, 92.91% accuracy was obtained to classify the six human activities. This performance is because ICEEMDAN allowed the extraction of information that was embedded in the signal, and the segmentation of the IMF allowed the network to discriminate between static and dynamic activities.

10

Development of an AI-based audiogram classification method for patient referral

80%

Kassjański M. , Kulawiak M. , Przewoźny T.

Annals of Computer Science and Information Systems

|

2022

|

tom Vol. 30

163--168

EN

Hearing loss is one of the most significant sensory disabilities. It can have various negative effects on a person's quality of life, ranging from impeded school and academic performance to total social isolation in severe cases. It is therefore vital that early symptoms of hearing loss are diagnosed quickly and accurately. Audiology tests are commonly performed with the use of tonal audiometry, which measures a patient's hearing threshold both in air and bone conduction at different frequencies. The graphic result of this test is represented on an audiogram, which is a diagram depicting the values of the patient's measured hearing thresholds. In the course of the presented work several different artificial neural network models, including MLP, CNN and RNN, have been developed and tested for classification of audiograms into two classes - normal and pathological represented hearing loss. The networks have been trained on a set of 2400 audiograms analysed and classified by professional audiologists. The best classification performance was achieved by the RNN architecture (represented by simple RNN, GRU and LSTM), with the highest out-of-training accuracy being 98% for LSTM. In clinical application, the developed classifier can significantly reduce the workload of audiology specialists by enabling the transfer of tasks related to analysis of hearing test results towards general practitioners. The proposed solution should also noticeably reduce the patient's average wait time between taking the hearing test and receiving a diagnosis. Further work will concentrate on automating the process of audiogram interpretation for the purpose of diagnosing different types of hearing loss.

11

Influence of modelling phase transformations with the use of LSTM network on the accuracy of computations of residual stresses for the hardening process

80%

Wróbel J. , Kulawik A.

Bulletin of the Polish Academy of Sciences. Technical Sciences

|

2023

|

tom Vol. 71, nr 4

art. no. e145681

EN

Replacing mathematical models with artificial intelligence tools can play an important role in numerical models. This paper analyses the modeling of the hardening process in terms of temperature, phase transformations in the solid state and stresses in the elastic-plastic range. Currently, the use of artificial intelligence tools is increasing, both to make greater generalizations and to reduce possible errors in the numerical simulation process. It is possible to replace the mathematical model of phase transformations in the solid state with an artificial neural network (ANN). Such a substitution requires an ANN network that converts time series (temperature curves) into shares of phase transformations with a small training error. With an insufficient training level of the network, significant differences in stress values will occur due to the existing couplings. Long-Short-Term Memory (LSTM) networks were chosen for the analysis. The paper compares the differences in stress levels with two coupled models using a macroscopic model based on CCT diagram analysis and using the Johnson-Mehl-Avrami-Kolmogorov (JMAK) and Koistinen-Marburger (KM) equations, against the model memorized by the LSTM network. In addition, two levels of network training accuracy were also compared. Considering the results obtained from the model based on LSTM networks, it can be concluded that it is possible to effectively replace the classical model in modeling the phenomena of the heat treatment process.

12

Detecting type of hearing loss with different AI classification methods: a performance review

80%

Kassjański M. , Kulawiak M. , Przewoźny T. , Tretiakow D. , Kuryłowicz J. , Molisz A. , Koźmiński K. , Kwaśniewska A. , Mierzwińska-Dolny P. , Grono M.

Annals of Computer Science and Information Systems

|

2023

|

tom Vol. 35

1017--1022

EN

Hearing is one of the most crucial senses for all humans. It allows people to hear and connect with the environment, the people they can meet and the knowledge they need to live their lives to the fullest. Hearing loss can have a detrimental impact on a person's quality of life in a variety of ways, ranging from fewer educational and job opportunities due to impaired communication to social withdrawal in severe situations. Early diagnosis and treatment can prevent most hearing loss. Pure tone audiometry, which measures air and bone conduction hearing thresholds at various frequencies, is widely used to assess hearing loss. A shortage of audiologists might delay diagnosis since they must analyze an audiogram, a graphic representation of pure tone audiometry test results, to determine hearing loss type and treatment. In the presented work, several AI-based models were used to classify audiograms into three types of hearing loss: mixed, conductive, and sensorineural. These models included Logistic Regression, Support Vector Machines, Stochastic Gradient Descent, Decision Trees, RandomForest, Feedforward Neural Network (FNN), Convolutional Neural Network (CNN), Graph Neural Network (GNN), and Recurrent Neural Network (RNN). The models were trained using 4007 audiograms classified by experienced audiologists. The RNN architecture achieved the best classification performance, with an out-of-training accuracy of 94.46%. Further research will focus on increasing the dataset and enhancing the accuracy of RNN models.

13

Forecasting future values of time series using the lstm network on the example of currencies and WIG20 companies

80%

Mróz B. , Nowicki F.

Zeszyty Naukowe. Telekomunikacja i Elektronika / Uniwersytet Technologiczno-Przyrodniczy w Bydgoszczy

|

2020

|

tom nr 24

13--30

EN

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.

PL

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.

14

Shallow, Deep, Ensemble models for Network Device Workload Forecasting

80%

Liu C.

Annals of Computer Science and Information Systems

|

2020

|

tom Vol. 21

101--104

EN

Reliable prediction of workload-related characteristics of monitored devices is important and helpful for management of infrastructure capacity. This paper presents 3 machine learning models (shallow, deep, ensemble) with different complexity for network device workload forecasting. The performance of these models have been compared using the data provided in FedCSIS'20 Challenge. The R2 scores achieved from the cascade Support Vector Regression (SVR) based shallow model, Long short-term memory (LSTM) based deep model, and hierarchical linear weighted ensemble model are 0.2506, 0.2831, and 0.3059, respectively, and was ranked 3rd place in the preliminary stage of the challenges.