• Journal of Broadcast Engineering
• /
• v.28 no.2
• /
• pp.238-241
• /
• 2023

Recently deep learning models are widely used for various tasks such as facial recognition and face editing. Their training process often involves a dataset with imbalanced age distribution. It is because some age groups (teenagers and middle age) are more socially active and tends to have more data compared to the less socially active age groups (children and elderly). This imbalanced age distribution may negatively impact the deep learning training process or the model performance when tested against those age groups with less data. To this end, we propose an age-controllable face synthesis technique using a feature decomposition to classify age from facial images which can be utilized to synthesize novel data to balance out the age distribution. We perform extensive qualitative and quantitative evaluation on our proposed technique using the FFHQ dataset and we show that our method has better performance than existing method.

• Journal of Intelligence and Information Systems
• /
• v.29 no.3
• /
• pp.125-144
• /
• 2023

Recently, the use of Large Language Models (LLMs) such as ChatGPT has been increasing in various fields such as interactive commerce and mobile financial services. However, LMMs, which are mainly created by learning existing documents, can also learn various human biases inherent in documents. Nevertheless, there have been few comparative studies on the aspects of bias and discrimination in LLMs. The purpose of this study is to examine the existence and extent of nine types of discrimination (Age, Disability status, Gender identity, Nationality, Physical appearance, Race ethnicity, Religion, Socio-economic status, Sexual orientation) in LLMs and suggest ways to improve them. For this purpose, we utilized BBQ (Bias Benchmark for QA), a tool for identifying discrimination, to compare three large-scale language models including ChatGPT, GPT-3, and Bing Chat. As a result of the evaluation, a large number of discriminatory responses were observed in the mega-language models, and the patterns differed depending on the mega-language model. In particular, problems were exposed in elder discrimination and disability discrimination, which are not traditional AI ethics issues such as sexism, racism, and economic inequality, and a new perspective on AI ethics was found. Based on the results of the comparison, this paper describes how to improve and develop large-scale language models in the future.

• The Mathematical Education
• /
• v.40 no.1
• /
• pp.93-102
• /
• 2001

The world we live in is called the age of information. Thus communication and computers are doing the central role in it. When one studies the mathematical problem, the use of tools such as computers, calculators and technology is available for all students, and then students are actively engaged in reasoning, communicating, problem solving, and making connections with mathematics, between mathematics and other disciplines. The use of technology extends to include computer algebra systems, spreadsheets, dynamic geometry software and the Internet and help active learning of students by analyzing data and realizing mathematical models visually. In this paper, we explain concepts of transformation, linear transformation, congruence transformation and homothety, and introduce interesting, meaningful and visual models for teaching of a plane transformation geomeoy which are obtained by using Mathematica. Moreover, this study will show how to visualize linear transformation for student's better understanding in teaching a plane transformation geometry in classroom. New development of these kinds of teaching-learning methods can simulate student's curiosity about mathematics and their interest. Therefore these models will give teachers the active teaching and also give students the successful loaming for obtaining the concept of linear transformation.

• The Bulletin of The Korean Astronomical Society
• /
• v.36 no.1
• /
• pp.59.2-59.2
• /
• 2011

Constraining physical (or stellar population) properties - such as stellar mass, star-formation rate, stellar population age, and dust-extinction - of galaxies from observation is crucial in the study of galaxy evolution. This is very challenging especially for high-redshift galaxies, and a widely-used method to estimate physical properties of high-redshift galaxies is to compare their photometric spectral energy distributions (SEDs) to spectral templates from stellar population synthesis models. I will show that the SED-fitting results of high-redshift galaxies are strongly dependent on the assumed forms of star-formation histories. I will also present the results of SED-fitting analysis of observed Lyman-break galaxies which show that parametric models with gradually increasing star-formation histories provide better estimates of physical parameters of high-redshift (z>3) star-forming galaxies than traditionally-used exponentially declining star-formation histories. This result is also consistent with the predictions from the modern galaxy formation models.

• Journal of The Korean Astronomical Society
• /
• v.52 no.5
• /
• pp.173-180
• /
• 2019

We present IR flux density measurements, models of the broadband SED, and results of SED modeling for the Pulsar Wind Nebula (PWN) 3C 58. We find that the Herschel flux density seems to be slightly lower than suggested by interpolation of previous measurements in nearby wavebands, implying that there may be multiple electron populations in 3C 58. We model the SED using a simple stationary one-zone and a more realistic time-evolving multi-zone scenario. The latter includes variations of flow properties in the PWN (injected energy, magnetic field, and bulk speed), radiative energy losses, adiabatic expansion, and diffusion, similar to previous PWN models. From the modeling, we find that a PWN age of 2900-5400 yrs is preferred and that there may be excess emission at ${\sim}10^{11}Hz$. The latter may imply multiple populations of electrons in the PWN.

• Computers and Concrete
• /
• v.18 no.2
• /
• pp.155-163
• /
• 2016

This paper, proposes 20 models for predicting compressive strength of recycled aggregate concrete (RAC) containing silica fume by using gene expression programming (GEP). To construct the models, experimental data of 228 specimens produced from 61 different mixtures were collected from the literature. 80% of data sets were used in the training phase and the remained 20% in testing phase. Input variables were arranged in a format of seven input parameters including age of the specimen, cement content, water content, natural aggregates content, recycled aggregates content, silica fume content and amount of superplasticizer. The training and testing showed the models have good conformity with experimental results for predicting the compressive strength of recycled aggregate concrete containing silica fume.

• Journal of The Korean Astronomical Society
• /
• v.56 no.1
• /
• pp.41-57
• /
• 2023

We measured temporal and emission properties of quiescent magnetars using archival Chandra and XMM-Newton data, produced a list of the properties for 17 magnetars, and revisited previously suggested correlations between the properties. Our studies carried out with a larger sample, better spectral characterizations, and more thorough analyses not only confirmed previously-suggested correlations but also found new ones. The observed correlations differ from those seen in other neutron-star populations but generally accord with magnetar models. Specifically, the trends of the intriguing correlations of blackbody luminosity (L_BB) with the spin-inferred dipole magnetic field strength (B_S) and characteristic age (τ_c) were measured to be L_BB ∝ B^1.5_S and L_BB ∝ τ^-0.6_c, supporting the twisted magnetosphere and magnetothermal evolution models for magnetars. We report the analysis results and discuss our findings in the context of magnetar models.

• Proceedings of the PSK Conference
• /
• 2003.04a
• /
• pp.84-87
• /
• 2003

The purposes of this study were to evaluate the population pharmacokinetics of levosulpiride and terbinafine according to several pharmacokinetic models and to investigate the influence of characteristics of subjects such as age, body weight, height and serum creatinine concentration on the pharmacokinetic parameters of levosulpiride and terbinafine, respectively. (omitted)

• Asian-Australasian Journal of Animal Sciences
• /
• v.21 no.7
• /
• pp.923-927
• /
• 2008

The objective of this study was to compare random regression model and multiple trait animal model estimates of the (co) variance of total sperm cells over the active lifetime of AI boars. Data were provided by Smithfield Premium Genetics (Rose Hill, NC). Total number of records and animals for the random regression model were 19,629 and 1,736, respectively. Data for multiple trait animal model analyses were edited to include only records produced at 9, 12, 15, 18, 21, 24, and 27 months of age. For the multiple trait method estimates of genetic and residual variance for total sperm cells were heterogeneous among age classifications. When comparing multiple trait method to random regression, heritability estimates were similar except for total sperm cells at 24 months of age. The multiple trait method also resulted in higher estimates of heritability of total sperm cells at every age when compared to random regression results. Random regression analysis provided more detail with regard to changes of variance components with age. Random regression methods are the most appropriate to analyze semen traits as they are longitudinal data measured over the lifetime of boars.

• Genomics & Informatics
• /
• v.15 no.4
• /
• pp.156-161
• /
• 2017

Transcriptome analysis has been widely used to make biomarker panels to diagnose cancers. In breast cancer, the age of the patient has been known to be associated with clinical features. As clinical transcriptome data have accumulated significantly, we classified all human genes based on age-specific differential expression between normal and breast cancer cells using public data. We retrieved the values for gene expression levels in breast cancer and matched normal cells from The Cancer Genome Atlas. We divided genes into two classes by paired t test without considering age in the first classification. We carried out a secondary classification of genes for each class into eight groups, based on the patterns of the p-values, which were calculated for each of the three age groups we defined. Through this two-step classification, gene expression was eventually grouped into 16 classes. We showed that this classification method could be applied to establish a more accurate prediction model to diagnose breast cancer by comparing the performance of prediction models with different combinations of genes. We expect that our scheme of classification could be used for other types of cancer data.