• Journal of The Korean Society of Agricultural Engineers
• /
• v.63 no.1
• /
• pp.11-25
• /
• 2021

The objective of this study was to analyze the impact of activation functions on flood forecasting model based on Artificial neural networks (ANNs). The traditional activation functions, the sigmoid and tanh functions, were compared with the functions which have been recently recommended for deep neural networks; the ReLU, leaky ReLU, and ELU functions. The flood forecasting model based on ANNs was designed to predict real-time runoff for 1 to 6-h lead time using the rainfall and runoff data of the past nine hours. The statistical measures such as R2, Nash-Sutcliffe Efficiency (NSE), Root Mean Squared Error (RMSE), the error of peak time (ETp), and the error of peak discharge (EQp) were used to evaluate the model accuracy. The tanh and ELU functions were most accurate with R2=0.97 and RMSE=30.1 (㎥/s) for 1-h lead time and R2=0.56 and RMSE=124.6~124.8 (㎥/s) for 6-h lead time. We also evaluated the learning speed by using the number of epochs that minimizes errors. The sigmoid function had the slowest learning speed due to the 'vanishing gradient problem' and the limited direction of weight update. The learning speed of the ELU function was 1.2 times faster than the tanh function. As a result, the ELU function most effectively improved the accuracy and speed of the ANNs model, so it was determined to be the best activation function for ANNs-based flood forecasting.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.22 no.10
• /
• pp.1020-1023
• /
• 2011

In this paper, we present a symbol remapping method of BRGC M-ary QAM signal by using LLR soft bit decision information which is obtained after iterative decoding process. In order to reconstruct estimated transmitted signal constellation, we have to use exponential or hyperbolic tangent(tanh) function resulting in high implementation complexity. The BRGC mapping rule enables us to use a recursive operation. In addtion, we reduce the implementing complexity by using a curve fitting algorithm.

• Journal of Intelligence and Information Systems
• /
• v.10 no.2
• /
• pp.53-63
• /
• 2004

Recurrent-Cascade-Correlation(RCC) is a supervised teaming algorithm that automatically determines the size and topology of the network. RCC adds new hidden neurons one by one and creates a multi-layer structure in which each hidden layer has only one neuron. By second order RCC, new hidden neurons are added to only one hidden layer. These created neurons are not connected to each other. We present a generalization of the RCC Architecture by combining the standard RCC Architecture and the second order RCC Architecture. Whenever a hidden neuron has to be added, the new RCC teaming algorithm automatically determines whether the network topology grows vertically or horizontally. This new algorithm using sigmoid, tanh and new activation functions was tested with the morse-benchmark-problem. Therefore we recognized that the number of hidden neurons was decreased by the experiments of the RCC network generalization which used the activation functions.

• Journal of Korean Ophthalmic Optics Society
• /
• v.6 no.2
• /
• pp.17-21
• /
• 2001

We analyzed the luminance in the visual light region and the size of pupil by the luminance to estimate an UV-A line cut efficiency in the lens. The size of pupil by the luminance(L) was given by ${\Phi}=d-e{\cdot}tanh(f{\cdot}logL)$ and the transmittance efficiency value of a size of pupil was given by $T_r(r)=1-gr^2+hr^4$. We derived the absolute cut efficiency value ${\alpha}$ and the exclusion index $b=(1-{\alpha}){\times}100%$ about the UV-A in the $320{\sim}400nm$ regions. The ${\alpha}$ and b values were obtained respectively 0.018, 0.31, 0.273, 0.153 and 98, 69, 72, 85% of Uv-cut Lens, CR-39, red color and blue color.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.7A
• /
• pp.512-521
• /
• 2012

In this paper, we propose feasible combining rules for a decoding scheme of product codes to apply finite soft decision. Since the decoding scheme of product codes are based on complex tanh calculation with infinite soft decision, it requires high decoding complexity and is hard to practically implement. Thus, simple methods to construct look-up tables for finite soft decision are derived by analyzing the operations of the scheme. Moreover, we focus on using convolutional codes, which is popular for easy application of finite soft decision, as the horizontal codes of product codes so that the proposed decoding scheme can be properly implemented. Numerical results show that the performance of the product codes with convolutional codes using 4-bit soft decision approaches to that of same codes using infinite soft decision.

• Journal of the Korean Magnetics Society
• /
• v.24 no.1
• /
• pp.11-17
• /
• 2014

We measured the ferromagnetic resonance (FMR) signal using the monodisperse iron oxide nanoparticles with size D=4.67 nm, 5.64 nm and 6.34 nm synthesized by using the thermal decomposition method, respectively. The measured ferromagnetic resonance signals were compared with the calculated ones for superparamagnetic nanoparticles with lognormal volume distribution. The FMR linewidth broadening was propositional to tanh($V^2$), where V was volume of nanoparticles. The narrow linewidth of small size nanoparticles was due to the surface spins, while the broad linewidth of large size nanoparticles was due to the bulk spins affected by the crystalline structure of iron oxide nanoparticles. The superposition of surface and bulk effect was confirmed at D=5.64 nm nanoparticles, which was near the critical size for linewidth transition from surface effect to bulk effect.

• International Journal of Internet, Broadcasting and Communication
• /
• v.5 no.2
• /
• pp.5-10
• /
• 2013

We proposed a novel similarity measure using weighted difference entropy (WDE) to improve the performance of the CF system. The proposed similarity metric evaluates the entropy with a preference score difference between the common rated items of two users, and normalizes it based on the Gaussian, tanh and sigmoid function. We showed significant improvement of experimental results and environments. These experiments involved changing the number of nearest neighborhoods, and we presented experimental results for two data sets with different characteristics, and results for the quality of recommendation.

• ETRI Journal
• /
• v.34 no.4
• /
• pp.629-632
• /
• 2012

This letter investigates the combination of the Chase-2 and sum-product (SP) algorithms for low-density parity-check (LDPC) codes. A simple modification of the tanh rule for check node update is given, which incorporates test error patterns (TEPs) used in the Chase algorithm into SP decoding of LDPC codes. Moreover, a simple yet effective approach is proposed to construct TEPs for dealing with decoding failures with low-weight syndromes. Simulation results show that the proposed algorithm is effective in improving both the waterfall and error floor performance of LDPC codes.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 1999.04a
• /
• pp.23-23
• /
• 1999

후류손실을 가지는 혼합 전단층에 대하여 밀도변화가 없는 유동 및 밀도변화가 있는 유동의 선형 불안정성 해석을 수행하였다. 기본 유동의 속도장 및 밀도장은 tanh 함수를 사용하였으며, Gaussian 형태의 해석적 함수를 사용하여 두 유동을 분리시키는 평판 바로 다음에 존재하는 후류 손실 유동을 포함시켰다. 공간적 선형 불안정성 해석을 수행하여 불안정성 모드의 성장률과 파장속도를 주파수의 함수로서 구하였다. 해석 결과로부터 후류 손실을 가지는 혼합층은 sinuous 모드와 varicose 모드의 두 개의 불안정성 모드를 가짐을 알았다. 밀도가 균일한 경우에는 varicose 모드보다 sinuous 모드가 지배적이다. 밀도가 균일한 경우에는 varicose 모드보다 sinuous 모드가 지배적이다. 밀도구배가 존재하나 빠른 자유유동의 밀도가 높은 경우에는 밀도가 균일한 경우와 마찬가지로 sinuous 모드가 지배적인 모드가 된다. 그러나 느린 자유 유동의 밀도가 높은 경우에는 밀도장의 두께가 속도장의 두께보다 상대적으로 얇아지면 varicose 모드가 sinuous 모드보다 더욱 불안정하여질 수 있다. varicose 모드와 sinuous 모드의 성장률이 비슷한 밀도장의 두께에서는 두 불안정성 모드가 주파수 변화에 따라 분지 되어지는 경향을 보인다.

• International Journal of Internet, Broadcasting and Communication
• /
• v.12 no.3
• /
• pp.125-130
• /
• 2020

At the end of 1997, the volatility of the exchange rate intensified as the nation's exchange rate system was converted into a free-floating exchange rate system. As a result, managing the exchange rate is becoming a very important task, and the need for forecasting the exchange rate is growing. The exchange rate prediction model using the existing exchange rate prediction method, statistical technique, cannot find a nonlinear pattern of the time series variable, and it is difficult to analyze the time series with the variability cluster phenomenon. And as the number of variables to be analyzed increases, the number of parameters to be estimated increases, and it is not easy to interpret the meaning of the estimated coefficients. Accordingly, the exchange rate prediction model using artificial neural network, rather than statistical technique, is presented. Using DNN, which is the basis of deep learning among artificial neural networks, and LSTM, a recurrent neural network model, the number of hidden layers, neurons, and activation function changes of each model found the optimal exchange rate prediction model. The study found that although there were model differences, LSTM models performed better than DNN models and performed best when the activation function was Tanh.