• 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.

• Journal of the Society of Naval Architects of Korea
• /
• v.55 no.6
• /
• pp.466-473
• /
• 2018

Most frequency domain-based approaches assume that structural response should be a Gaussian random process. But a lot of non-Gaussian processes caused by multi-excitation and non-linearity in structural responses or load itself are observed in many real engineering problems. In this study, the effect of non-Normality on fatigue damages are discussed through case study. The accuracy of four frequency domain methods for non-Gaussian processes are compared in the case study. Power-law and Hermite models which are derived for non-Gaussian narrow-banded process tend to estimate fatigue damages less accurate than time domain results in small kurtosis and in case of large kurtosis they give conservative results. Weibull model seems to give conservative results in all environmental conditions considered. Among the four methods, Benascuitti-Tovo model for non-Gaussian process gives the best results in case study. This study could serve as background material for understanding the effect of non-normality on fatigue damages.

• 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.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.39 no.1
• /
• pp.64-72
• /
• 2016

In this paper, we utilize a Gaussian process to predict the power consumption in the air-conditioning system. As the power consumption in the air-conditioning system takes a form of a time-series and the prediction of the power consumption becomes very important from the perspective of the efficient energy management, it is worth to investigate the time-series model for the prediction of the power consumption. To this end, we apply the Gaussian process to predict the power consumption, in which the Gaussian process provides a prior probability to every possible function and higher probabilities are given to functions that are more likely consistent with the empirical data. We also discuss how to estimate the hyper-parameters, which are parameters in the covariance function of the Gaussian process model. We estimated the hyper-parameters with two different methods (marginal likelihood and leave-one-out cross validation) and obtained a model that pertinently describes the data and the results are more or less independent of the estimation method of hyper-parameters. We validated the prediction results by the error analysis of the mean relative error and the mean absolute error. The mean relative error analysis showed that about 3.4% of the predicted value came from the error, and the mean absolute error analysis confirmed that the error in within the standard deviation of the predicted value. We also adopt the non-parametric Wilcoxon's sign-rank test to assess the fitness of the proposed model and found that the null hypothesis of uniformity was accepted under the significance level of 5%. These results can be applied to a more elaborate control of the power consumption in the air-conditioning system.

• Bulletin of the Korean Mathematical Society
• /
• v.40 no.2
• /
• pp.269-279
• /
• 2003

We prove an empirical LIL for the Kaplan-Meier integral process constructed from the random censorship model under bracketing entropy and mild assumptions due to censoring effects. The main method in deriving the empirical LIL is to use a weak convergence result of the sequential Kaplan-Meier integral process whose proofs appear in Bae and Kim [2]. Using the result of weak convergence, we translate the problem of the Kaplan Meier integral process into that of a Gaussian process. Finally we derive the result using an empirical LIL for the Gaussian process of Pisier [6] via a method adapted from Ossiander [5]. The result of this paper extends the empirical LIL for IID random variables to that of a random censorship model.

• IEIE Transactions on Smart Processing and Computing
• /
• v.4 no.4
• /
• pp.202-208
• /
• 2015

In this study, we propose a new inference algorithm for a multiclass Gaussian process classification model using a variational EM framework and the Laplace approximation (LA) technique. This is performed in two steps, called expectation and maximization. First, in the expectation step (E-step), using Bayes' theorem and the LA technique, we derive the approximate posterior distribution of the latent function, indicating the possibility that each observation belongs to a certain class in the Gaussian process classification model. In the maximization step, we compute the maximum likelihood estimators for hyper-parameters of a covariance matrix necessary to define the prior distribution of the latent function by using the posterior distribution derived in the E-step. These steps iteratively repeat until a convergence condition is satisfied. Moreover, we conducted the experiments by using synthetic data and Iris data in order to verify the performance of the proposed algorithm. Experimental results reveal that the proposed algorithm shows good performance on these datasets.

• The Bulletin of The Korean Astronomical Society
• /
• v.45 no.1
• /
• pp.69.1-69.1
• /
• 2020

While Dark Energy is one of the explanations for the accelerating expansion of the Universe, its nature remains a mystery. The standard (flat) ΛCDM model is consistent with cosmological observations: type Ia Supernova, BAO, CMB, and so on. However, the analysis of observations assuming a model, model-dependent approach, is likely to bias the results towards the assumed model. In this poster, I will introduce model-independent approach with Gaussian process and the application of Gaussian process regression to reconstruct the equation of state of dark energy.

• Journal of the Korean Data and Information Science Society
• /
• v.27 no.3
• /
• pp.803-814
• /
• 2016

Recent times have seen an exponential increase in the amount of spatial data, which is in many cases associated with temporal data. Recent advances in computer technology and computation of hierarchical Bayesian models have enabled to analyze complex spatio-temporal data. Our work aims at modeling data of daily average nitrogen dioxide (NO2) levels obtained from 25 air monitoring sites in Seoul between 2003 and 2010. We considered an independent Gaussian process model and an auto-regressive model and carried out estimation within a hierarchical Bayesian framework with Markov chain Monte Carlo techniques. A Gaussian predictive process approximation has shown the better prediction performance rather than a Hierarchical auto-regressive model for the illustrative NO2 concentration levels at any unmonitored location.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.12 no.1
• /
• pp.755-767
• /
• 2020

A significant development has been made on a new fatigue damage model applicable to Gaussian wide band stress response spectra using numerical approximation methods such as data processing, time simulation, and regression analysis. So far, most of the alternative approximate models provide slightly underestimated or overestimated damage results compared with the rain-flow counting distribution. A more reliable approximate model that can minimize the damage differences between exact and approximate solutions is required for the practical design of ships and offshore structures. The present paper provides a detailed description of the development process of a new fatigue damage model. Based on the principle of the Gaussian wide band model, this study aims to develop the best approximate fatigue damage model. To obtain highly accurate damage distributions, this study deals with some prominent research findings, i.e., the moment of rain-flow range distribution MRR(n), the special bandwidth parameter μk, the empirical closed form model consisting of four probability density functions, and the correction factor QC. Sequential prerequisite data processes, such as creation of various stress spectra, extraction of stress time history, and the rain-flow counting stress process, are conducted so that these research findings provide much better results. Through comparison studies, the proposed model shows more reliable and accurate damage distributions, very close to those of the rain-flow counting solution. Several significant achievements and findings obtained from this study are suggested. Further work is needed to apply the new developed model to crack growth prediction under a random stress process in view of the engineering critical assessment of offshore structures. The present developed formulation and procedure also need to be extended to non-Gaussian wide band processes.

• Wind and Structures
• /
• v.36 no.1
• /
• pp.15-29
• /
• 2023

To better estimate the non-Gaussian extreme wind pressures for high-rise buildings, a data-driven revised Hermitetype peak factor estimation model is proposed in this papar. Subsequently, a comparative study on three types of methods, such as Hermite-type models, short-time estimate Gumbel method (STE), and new translated-peak-process method (TPP) is carried out. The investigations show that the proposed Hermite-type peak factor has better accuracy and applicability than the other Hermite-type models, and its absolute accuracy is slightly inferior to the STE and new TPP methods for non-Gaussian wind pressures by comparing with the observed values. Moreover, these methods generally overestimate the Gaussian wind pressures especially the STE.