• Journal of KIISE
• /
• v.45 no.2
• /
• pp.157-164
• /
• 2018

Due to the revolutionary advances in deep learning, performance of pattern recognition has increased significantly in many applications like speech recognition and image recognition, and some systems outperform human-level intelligence in specific domains. Unlike pattern recognition, in this paper, we focus on generating Korean sentences based on a few Korean sentences. We apply variational recurrent auto-encoder (VRAE) and modify the model considering some characteristics of Korean sentences. To reduce the number of words in the model, we apply a word spacing model. Also, there are many Korean sentences which have the same meaning but different word order, even without subjects or objects; therefore we change the unidirectional encoder of VRAE into a bidirectional encoder. In addition, we apply an interpolation method on the encoded vectors from the given sentences, so that we can generate new sentences which are similar to the given sentences. In experiments, we confirm that our proposed method generates better sentences which are semantically more similar to the given sentences.

• KSCE Journal of Civil and Environmental Engineering Research
• /
• v.40 no.2
• /
• pp.233-238
• /
• 2020

Most deep learning (DL) approaches for bridge damage localization based on a structural health monitoring system commonly use supervised learning-based DL models. The supervised learning-based DL model requires the response data obtained from sensors on the bridge and also the label which indicates the damaged state of the bridge. However, it is impractical to accurately obtain the label data in fields, thus, the supervised learning-based DL model has a limitation in that it is not easily applicable in practice. On the other hand, an unsupervised learning-based DL model has the merit of being able to train without label data. Considering this advantage, this study aims to propose and theoretically validate a damage localization approach for bridges using a variational autoencoder, a representative unsupervised learning-based DL network: as a result, this study indicated the feasibility of VAE for damage localization.

• Journal of Broadcast Engineering
• /
• v.28 no.3
• /
• pp.275-284
• /
• 2023

Gene expression data can be utilized in various studies, including the prediction of disease prognosis. However, there are challenges associated with collecting enough data due to cost constraints. In this paper, we propose a gene expression data generation model based on Conditional Variational Autoencoder. Our results demonstrate that the proposed model generates synthetic data with superior quality compared to two other state-of-the-art models for gene expression data generation, namely the Wasserstein Generative Adversarial Network with Gradient Penalty based model and the structured data generation models CTGAN and TVAE.

• Journal of Welding and Joining
• /
• v.18 no.4
• /
• pp.55-63
• /
• 2000

This paper presents a study of the solder bumping process. The theoretical model, based on the variational principle instead of solving the Navier-Stokes equation with moving boundaries, was developed to considered the energy dissipation in semi-solid phase and the approximate solidification time of the molten metal droplet. The simulation results revealed that the developed model could reasonably describe the collision behavior of molten metal with solid surface. Simulations were made with variation of initial droplet temperature, substrate metal and initial substrate temerature.

• 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 Journal of the Acoustical Society of Korea
• /
• v.40 no.6
• /
• pp.578-586
• /
• 2021

We propose a semi-supervised learning method based on Variational AutoEncoder (VAE) and Unsupervised Data Augmentation (UDA) to improve the performance of an end-to-end speech recognizer. In the proposed method, first, the VAE-based augmentation model and the baseline end-to-end speech recognizer are trained using the original speech data. Then, the baseline end-to-end speech recognizer is trained again using data augmented from the learned augmentation model. Finally, the learned augmentation model and end-to-end speech recognizer are re-learned using the UDA-based semi-supervised learning method. As a result of the computer simulation, the augmentation model is shown to improve the Word Error Rate (WER) of the baseline end-to-end speech recognizer, and further improve its performance by combining it with the UDA-based learning method.

• The Korean Journal of Applied Statistics
• /
• v.35 no.3
• /
• pp.445-455
• /
• 2022

In this paper, we introduce existing Bayesian methods for high-dimensional sparse regression models and compare their performance in various simulation scenarios. Especially, we focus on the variational Bayes approach proposed by Ray and Szabó (2021), which enables scalable and accurate Bayesian inference. Based on simulated data sets from sparse high-dimensional linear regression models, we compare the variational Bayes approach with other Bayesian and frequentist methods. To check the practical performance of the variational Bayes in logistic regression models, a real data analysis is conducted using leukemia data set.

• Journal of Power System Engineering
• /
• v.19 no.3
• /
• pp.55-62
• /
• 2015

This paper discusses a new model for investigating the micro-mechanical behavior of beam-like structures composed of various elastic moduli and complex geometries varying through the cross-sectional directions and also periodically-repeated along the axial directions. The original three-dimensional problem is first formulated in an unified and compact intrinsic form using the concept of decomposition of the rotation tensor. Taking advantage of two smallness of the cross-sectional dimension-to-length parameter and the micro-to-macro heterogeneity and performing homogenization along dimensional reduction simultaneously, the variational asymptotic method is used to rigorously construct an effective zeroth-order beam model, which is similar a generalized Timoshenko one (the first-order shear deformation model) capable of capturing the transverse shear deformations, but still carries out the zeroth-order approximation which can maximize simplicity and promote efficiency. Two examples available in literature are used to demonstrate the consistence and efficiency of this new model, especially for the structures, in which the effects of transverse shear deformations are significant.

• Journal of the Korean Society for Industrial and Applied Mathematics
• /
• v.26 no.3
• /
• pp.156-184
• /
• 2022

In this article, we propose a novel variational model for restoring color images corrupted by mixed multiplicative Gamma noise and additive Gaussian noise. The model involves a data-fidelity term that characterizes the mixed noise as an infimal convolution of two noise distributions and the saturation-value total variation (SVTV) regularization. The data-fidelity term facilitates suitable separation of the multiplicative Gamma and Gaussian noise components, promoting simultaneous elimination of the mixed noise. Furthermore, the SVTV regularization enables adequate denoising of homogeneous regions, while maintaining edges and details and diminishing the color artifacts induced by noise. To solve the proposed nonconvex model, we exploit an alternating minimization approach, and then the alternating direction method of multipliers is adopted for solving subproblems. This contributes to an efficient iterative algorithm. The experimental results demonstrate the superior performance of the proposed model compared to other existing or related models, with regard to visual inspection and image quality measurements.

• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.1
• /
• pp.11-18
• /
• 2022

De novo drug design is the process of developing new drugs that can interact with biological targets such as protein receptors. Traditional process of de novo drug design consists of drug candidate discovery and drug development, but it requires a long time of more than 10 years to develop a new drug. Deep learning-based methods are being studied to shorten this period and efficiently find chemical compounds for new drug candidates. Many existing deep learning-based drug design models utilize recurrent neural networks to generate a chemical entity represented by SMILES strings, but due to the disadvantages of the recurrent networks, such as slow training speed and poor understanding of complex molecular formula rules, there is room for improvement. To overcome these shortcomings, we propose a deep learning model for SMILES string generation using variational autoencoders with self-attention mechanism. Our proposed model decreased the training time by 1/26 compared to the latest drug design model, as well as generated valid SMILES more effectively.