• The Journal of the Korea Contents Association
• /
• v.21 no.3
• /
• pp.616-625
• /
• 2021

Deep neural networks are an approximation method that approximates an arbitrary function to a linear model and then repeats additional approximation using a nonlinear active function. In this process, the method of evaluating the performance of approximation uses the loss function. Existing in-depth learning methods implement approximation that takes into account loss functions in the linear approximation process, but non-linear approximation phases that use active functions use non-linear transformation that is not related to reduction of loss functions of loss. This study proposes parametric activation functions that introduce scale parameters that can change the scale of activation functions and location parameters that can change the location of activation functions. By introducing parametric activation functions based on scale and location parameters, the performance of nonlinear approximation using activation functions can be improved. The scale and location parameters in each hidden layer can improve the performance of the deep neural network by determining parameters that minimize the loss function value through the learning process using the primary differential coefficient of the loss function for the parameters in the backpropagation. Through MNIST classification problems and XOR problems, parametric activation functions have been found to have superior performance over existing activation functions.

• Journal of the Korean Society for Railway
• /
• v.16 no.5
• /
• pp.372-378
• /
• 2013

This paper presents an approximate model for computing the peak demand power in a metro railway system. The peak demand of substations can be calculated using the current vector iteration method. But the existing method requires many repeated calculations to determine the peak demand power, which makes it difficult to apply to the real-time peak power control problem. In this paper, we assume that none of the conditions vary except source impedance and make an approximate model for rapid calculation based on changes in the impedance of the power substation. The proposed model result is approximately the same as the existing model, which is demonstrated through simulation.

• The Korean Journal of Applied Statistics
• /
• v.24 no.2
• /
• pp.347-357
• /
• 2011

In this paper, we propose an approximation to the overshoot in M/$E_n$/1 queues. Overshoot means the size of excess over the threshold when the workload process of an M/$E_n$/1 queue exceeds a prespecified threshold. The distribution, $1^{st}$ and $2^{nd}$ moments of overshoot have an important role in solving some kind of optimization problems. For the approximation to the overshoot, we propose a formula that is a convex sum of the service time distribution and an exponential distribution. We also do a numerical study to check how exactly the proposed formula approximates the overshoot.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.8 no.8
• /
• pp.1235-1240
• /
• 2013

We investigate an approximate analytic expression of signal distortion caused by the interaction between self-phase modulation (SPM) and dispersion in the dispersion-managed optical transmission where the dispersion is optimally compensated. From the analytic study, we obtain the analytic expression of the signal distortion in dispersion-managed optical transmission system. To confirm the validity of the analytic expression, we show that the eye-opening penalties calculated by the analytic expression correspond with the simulation results. Using the analytic result, we can easily estimate the signal distortion without complex nonlinear simulations.

• Journal of the Korea Computer Industry Society
• /
• v.4 no.4
• /
• pp.443-448
• /
• 2003

Approximate k-nearest neighbor queries are frequently occurred for finding the k nearest neighbors to a given query point in spatial database systems. The number of searched nodes in an index must be minimized in order to increase the performance of approximate k nearest neighbor queries. In this paper. we suggest the technique of approximate k nearest neighbor queries on R-tree family by improving the existing algorithm and evaluate the performance of the proposed method in dynamic spatial database environments. The simulation results show that a proposed method always has a low number of disk access irrespective of object distribution, size of nearest neighbor queries and approximation rates as compared with an existing method.

• Journal of KIISE:Databases
• /
• v.27 no.2
• /
• pp.199-208
• /
• 2000

The k-nearest neighbor search query based on similarity is very important for content-based multimedia information retrieval(MIR). The conventional exact k-nearest neighbor search algorithm is not efficient for the MIR application because multimedia data should be represented as high dimensional feature vectors. Thus, an approximate k-nearest neighbor search algorithm is required for the MIR applications because the performance increase may outweigh the drawback of receiving approximate results. For this, we propose a new approximate k-nearest neighbor search algorithm for high dimensional data. In addition, the comparison of the conventional algorithm with our approximate k-nearest neighbor search algorithm is performed in terms of retrieval performance. Results show that our algorithm is more efficient than the conventional ones.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.3 s.345
• /
• pp.15-20
• /
• 2006

A Newton-Raphson Method for table driven algorithm is presented in this paper. We concentrate the approximation of square root by using Newton-Raphson method. We confirm that this method has advantages of accurate and fast processing with optimized initial point. Hence the selection of the fitted initial points used in approximation of Newton-Raphson algorithm is important issue. This paper proposes that log scale based on geometric wean is most profitable initial point. It shows that the proposed method givemore accurate results with faster processing speed.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.7 no.3
• /
• pp.395-402
• /
• 2003

The capacity of a CDMA cellular system is determined by the amount of interference, therefore the exact estimation of interference is important to evaluate the system performance. In this paper, we propose an approximated equation which calculates reverse link other cell interference in the CDMA cellular systems. The equation using Riemann-Zeta function has a property that it is useful in case of any radio propagation loss exponents. And we compare calculation results with simulation results in other to verify it's usefulness. The upper bound of system capacity calculated with the proposed approximated equation gives almost alike result with the simulation. The proposed interference bound is useful to calculate system capacity and to evaluate some algorithm in a hierarchical cellular system which must be considered various propagation exponents.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.35 no.11A
• /
• pp.1027-1037
• /
• 2010

In this paper, we propose the linear interpolation-based BD (Block Diagonalization) precoding approximation algorithm for low complexity in downlink multiuser MIMO-OFDM (Multiple-input Multiple-output Orthogonal Frequency Division Multiplexing) systems. In the case of applying the general BD precoding algorithm to multiuser MIMO-OFDM systems, the computational complexity increases in proportional to the number of subcarriers. The proposed interpolation-based BD precoding approximation algorithm can be achieved similar SER performance with general BD algorithm and can decrease the computational complexity. It is proved that proposed algorithm can achieve the significantly decreased computational complexity by computer simulation.

• Journal of KIISE:Computing Practices and Letters
• /
• v.16 no.6
• /
• pp.742-746
• /
• 2010

In this paper, we propose new randomness testing methods based on approximate periods in order to improve the previous randomness testing method using exact pattern matching. Finding approximate periods of random sequences enables us to search similarly repeated parts, but it has disadvantages since it takes long time. In this paper we propose randomness testing methods whose time complexity is O($n^2$) by reducing the time complexity of computing approximate periods from O($n^3$) to O($n^2$). Moreover, we perform some experiments to compare pseudo random number generated by AES cryptographic algorithms and true random number.