• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.49 no.2
• /
• pp.115-122
• /
• 2012

We present here a new class of activation functions for neural networks, which herein will be called CosGauss function. This function is a cosine-modulated gaussian function. In contrast to the sigmoidal-, hyperbolic tangent- and gaussian activation functions, more ridges can be obtained by the CosGauss function. It will be proven that this function can be used to aproximate polynomials and step functions. The CosGauss function was tested with a Cascade-Correlation-Network of the multilayer structure on the Tic-Tac-Toe game and iris plants problems, and results are compared with those obtained with other activation functions.

• Journal for History of Mathematics
• /
• v.35 no.1
• /
• pp.1-18
• /
• 2022

This paper presents a statistical machine learning method that generates the implied volatility surface under the rareness of the market data. We apply the practitioner's Black-Scholes model and Gaussian process regression method to construct a Bayesian inference system with observed volatilities as a prior information and estimate the posterior distribution of the unobserved volatilities. The variance instead of the volatility is the target of the estimation, and the radial basis function is applied to the mean and kernel function of the Gaussian process regression. We present two types of Gaussian process regression methods and empirically analyze them.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.13 no.4
• /
• pp.195-204
• /
• 2018

Lane keeping assist and departure prevention system are the key functions of ADAS. In this paper, we propose lane detection method which uses Gaussian function based RANSAC. The proposed method consists mainly of IPM (inverse perspective mapping), Canny edge detector, and Gaussian function based RANSAC (Random Sample Consensus). The RANSAC uses Gaussian function to extract the parameters of straight or curved lane. The proposed RANSAC is different from the conventional one, in the following two aspects. One is the selection of sample with different probability depending on the distance between sample and camera. Another is the inlier sample score that assigns higher weights to samples near to camera. Through simulations, we show that the proposed method can achieve good performance in various of environments.

• Korean Journal of Optics and Photonics
• /
• v.15 no.3
• /
• pp.200-213
• /
• 2004

The realization of partially coherent Gaussian Schell-model beams with isotropic cross section and anisotropic degree of coherence function is investigated theoretically. An optical system is devised to transform diffused light generated by passing the Gaussian beam of the He-Ne laser thorough a rotating holographic diffuser to the partially coherent Gaussian Schell-model beam with isotropic cross section and anisotropic degree of coherence function. Analytic design equations are formulated and design examples are presented.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.18 no.5
• /
• pp.1143-1148
• /
• 2014

This paper has analyzed the change of subthreshold swing for doping distribution function of asymmetric double gate(DG) MOSFET. The basic factors to determine the characteristics of DGMOSFET are dimensions of channel, i.e. channel length and channel thickness, and doping distribution function. The doping distributions are determined by ion implantation used for channel doping, and follow Gaussian distribution function. Gaussian function has been used as carrier distribution in solving the Poisson's equation. Since the Gaussian function is exactly not symmetric for top and bottome gates, the subthreshold swings are greatly changed for channel length and thickness, and the voltages of top and bottom gates for asymmetric double gate MOSFET. The deviation of subthreshold swings has been investigated for parameters of Gaussian distribution function such as projected range and standard projected deviation in this paper. As a result, we know the subthreshold swing is greatly changed for doping profiles and bias voltage.

• Proceedings of the Korean Statistical Society Conference
• /
• 2000.11a
• /
• pp.115-118
• /
• 2000

In this paper we establish limit results on the increments of (N, d)-Gaussian processes with independent components, via estimating upper bounds of large deviation probabilities on the suprema of (N, d)-Gaussian processes.

• Communications of the Korean Mathematical Society
• /
• v.13 no.4
• /
• pp.865-874
• /
• 1998

A modulus of continuity on the increments of a two-parameter Gaussian process is obtained via estimating large deviation probability inequalities on the suprema of the Gaussian process.

• Science of Emotion and Sensibility
• /
• v.17 no.2
• /
• pp.63-76
• /
• 2014

Although most behavioral reaction times (RTs) for cognitive tasks exhibit positively skewed distributions, the majority of studies primarily rely on a measure of central tendency (e.g. mean) which can cause misinterpretations of data's underlying property. The purpose of current study is to introduce procedures for describing characteristics of RT distributions, thereby effectively examine the influence of experimental manipulations. On the basis of assumption that RT distribution can be represented as a convolution of Gaussian and exponential variables, we fitted the ex-Gaussian function under a maximum-likelihood method. The ex-Gaussian function provides quantitative parameters of distributional properties and the probability density functions. Here we exemplified distributional analysis by using empirical RT data from two conventional visual search tasks, and attempted theoretical interpretation for setsize effect leading proportional mean RT delays. We believe that distributional RT analysis with a mathematical function beyond the central tendency estimates could provide insights into various theoretical and individual difference studies.

• Geomechanics and Engineering
• /
• v.30 no.4
• /
• pp.325-336
• /
• 2022

This paper analyzes the ground surface settlements induced by side-by-side twin shield tunnels bored in sedimentary soils, which primarily consist of sand with clay strata above the tunnel crown. The measurements were obtained during the construction of twin tunnels underneath the Incheon International Airport (IIA) located in Korea. The measured surface settlement troughs are approximated with Gaussian functions. The trough width parameters i and K of the settlement troughs produced by the first and second tunnel passings are determined, along with those for the total settlement trough. The surface settlement troughs produced by the first shield passing are reasonably represented by a symmetric Gaussian curve. The surface settlement troughs induced by the second shield tunnel display marginal asymmetric shapes at selected sections. The total settlement troughs are fitted both with a shifted symmetric Gaussian function and the superposition method utilizing an asymmetric function for the incremental trough produced by the second tunnel. It is revealed that the superposition method does not always produce better fits with the total settlement. Instead, the shifted symmetric Gaussian function is overall demonstrated to provide more favorable agreements with the recordings. Therefore, the shifted symmetric Gaussian function is recommended to be used in the design for the prediction of the settlement in clays caused by twin tunneling considering the simplicity of the procedure compared with the superposition method. The amount of increase in the width parameter K for the twin tunnel relative to that for the single tunnel is quantified, which can be used for a preliminary estimate of the surface settlement in clay induced by twin shield tunnels.

• Advances in nano research
• /
• v.5 no.1
• /
• pp.13-25
• /
• 2017

We examine the total optical dielectric function (TODF) of impurity doped GaAs quantum dot (QD) from the viewpoint of anisotropy, position-dependent effective mass (PDEM) and position dependent dielectric screening function (PDDSF), both in presence and absence of noise. The dopant impurity potential is Gaussian in nature and noise employed is Gaussian white noise that has been applied to the doped system via two different modes; additive and multiplicative. A change from fixed effective mass and fixed dielectric constant to those which depend on the dopant coordinate manifestly affects TODF. Presence of noise and also its mode of application bring about more rich subtlety in the observed TODF profiles. The findings indicate promising scope of harnessing the TODF of doped QD systems through expedient control of site of dopant incorporation and application of noise in desired mode.