• Journal of KIISE:Software and Applications
• /
• v.35 no.7
• /
• pp.417-423
• /
• 2008

In this paper we present a probabilistic method for source separation in the case here each source has a certain temporal structure. We tackle the problem of source separation by maximum pseudo-likelihood estimation, representing the latent function which characterizes the temporal structure of each source by a random process with a Gaussian prior. The resulting pseudo-likelihood of the data is Gaussian, determined by a mixing matrix as well as by the predictive mean and covariance matrix that can easily be computed by Gaussian process (GP) regression. Gradient-based optimization is applied to estimate the demixing matrix through maximizing the log-pseudo-likelihood of the data. umerical experiments confirm the useful behavior of our method, compared to existing source separation methods.

• Journal of the Korean Society of Industry Convergence
• /
• v.18 no.4
• /
• pp.191-196
• /
• 2015

In this paper, we present a path-smoothing algorithm for a robot arm manipulator that finds the path using a joint space-based rapidly-exploring random tree. Unlike other smoothing algorithms which require complex mathematical computation, the proposed path-smoothing algorithm is done using a Gaussian process. To find the optimal hyperparameters of the Gaussian process, we use differential evolution hybridized with opposition-based learning. The simulation result indicates that the Gaussian process whose hyperparameters were optimized by hybrid differential evolution successfully smoothed the path generated by the joint space-based rapidly-exploring random tree.

• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.8
• /
• pp.4285-4299
• /
• 2019

Adversarial attack is a technique that causes a malfunction of classification models by adding noise that cannot be distinguished by humans, which poses a threat to a deep learning model. In this paper, we propose an efficient method to detect adversarial images using Gaussian process regression. Existing deep learning-based adversarial detection methods require numerous adversarial images for their training. The proposed method overcomes this problem by performing classification based on the statistical features of adversarial images and clean images that are extracted by Gaussian process regression with a small number of images. This technique can determine whether the input image is an adversarial image by applying Gaussian process regression based on the intermediate output value of the classification model. Experimental results show that the proposed method achieves higher detection performance than the other deep learning-based adversarial detection methods for powerful attacks. In particular, the Gaussian process regression-based detector shows better detection performance than the baseline models for most attacks in the case with fewer adversarial examples.

• The Korean Journal of Applied Statistics
• /
• v.26 no.3
• /
• pp.389-399
• /
• 2013

This article introduces Gaussian process regression and shows its application with time-course microarray gene expression data. Gene screening for yeast cell cycle microarray expression data is accomplished with a ratio of log marginal likelihood that uses Gaussian process regression with a squared exponential covariance kernel function. Gaussian process regression fitting with each gene is done and shown with the nine top ranking genes. With the screened data the Gaussian model-based clustering is done and its silhouette values are calculated for cluster validity.

• Wind and Structures
• /
• v.35 no.6
• /
• pp.419-430
• /
• 2022

In wind-resistant designs, wind velocity is assumed to be a Gaussian process; however, local complex topography may result in strong non-Gaussian wind features. This study investigates the non-Gaussian wind features over complex terrain under atmospheric turbulent boundary layers by the large eddy simulation (LES) model, and the turbulent inlet of LES is generated by the consistent discretizing random flow generation (CDRFG) method. The performance of LES is validated by two different complex terrains in Changsha and Mianyang, China, and the results are compared with wind tunnel tests and onsite measurements, respectively. Furthermore, the non-Gaussian parameters, such as skewness, kurtosis, probability curves, and gust factors, are analyzed in-depth. The results show that the LES method is in good agreement with both mean and turbulent wind fields from wind tunnel tests and onsite measurements. Wind fields in complex terrain mostly exhibit a left-skewed Gaussian process, and it changes from a softening Gaussian process to a hardening Gaussian process as the height increases. A reduction in the gust factors of about 2.0%-15.0% can be found by taking into account the non-Gaussian features, except for a 4.4% increase near the ground in steep terrain. This study can provide a reference for the assessment of extreme wind loads on structures in complex terrain.

• Journal of the Korean Society for Aeronautical & Space Sciences
• /
• v.48 no.3
• /
• pp.175-185
• /
• 2020

The six-DOF aerodynamic coefficients of a missile entail inherent physical gradient constraints originated from the geometric characteristics of a cylindrical fuselage. To effectively adopt the freely available gradient information in aerodynamic coefficients modeling, this research employed gradient-enhanced (GE) Gaussian process. To investigate the accuracy of aerodynamic coefficients predicted with gradients information, we compared two Gaussian-process-based models: ordinary and GE Gaussian process models with and without gradient information, respectively. As a result, we found that GE Gaussian process models were able to comply with imposed gradient information and more accurate than ordinary Gaussian process models. However, we also found that GE Gaussian process modeling cannot handle gradient information continuously and ends up with more samples due to additional gradient information.

• The Journal of Korea Robotics Society
• /
• v.17 no.3
• /
• pp.296-302
• /
• 2022

This paper proposes a center of plantar foot pressure (CoP) trajectory estimation method based on Gaussian process regression, with the aim to show robust results regardless of the regions and numbers of FSRs of the insole sensor. This method can bring an interpolation between the measurement points inside the wearable insole sensor, and two experiments are conducted for performance evaluation. For this purpose, the input data used in the experiment are generated in three types (13 FSRs, 8 FSRs, 5 FSRs) according to the regions and numbers of FSRs. First, the estimation results of the CoP trajectory are compared using Gaussian process regression and weighted mean. As a result of each method, the estimation results of the two methods were similar in the case of 13 FSRs data. On the other hand, in the case of the 8 and 5 FSRs data, the weighted mean varies depending on the regions and numbers of FSRs, but the estimation results of Gaussian process regression showed similar results in spite of reducing the regions and numbers. Second, the estimation results of the CoP trajectory based on Gaussian process regression during several gait cycles are analyzed. In five gait cycles, the previous cycle and the current estimation results are compared, and it was confirmed that similar trajectories appeared in all. In this way, the method of estimating the CoP trajectory based on Gaussian process regression showed robust results, and stability was confirmed by yielding similar results in several gait cycles.

• Journal of applied mathematics & informatics
• /
• v.20 no.1_2
• /
• pp.585-594
• /
• 2006

We consider a stochastic integral process represented by multiple Ito-Wiener integrals. We derive gaussian chaos which has some shift continuous function. We get continuity property of self-similar process represented by multiple integrals and finally we show that $Y_{H_t}$ (t) is continuous in t with probability one for Holder function $H_t$ of exponent $\beta$.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.25 no.6
• /
• pp.1-6
• /
• 2002

Most of process data control have been designed under the assumption that there are independence between observed data. However, it has been difficult to apply the traditional method to realtime data because they are autocorrelated, and they are not normally distributed. And the more, they have fluctuating means. Already the control method for these data was proposed by Markov Modulated Gaussian Process. Therefore, this study take into account MMGP's traits especially for the MMGP's autocorrelation.