https://scholar.korea.ac.kr/handle/2021.sw.korea/267 Wed, 08 Apr 2026 12:06:35 GMT 2026-04-08T12:06:35Z https://scholar.korea.ac.kr/handle/2021.sw.korea/281149 Title: PyTorch로 배우는 딥러닝과 생성형 AI : 개념부터 vision, 생성형 모형, foundation 모형의 구현까지 Authors: Park, Yousung Sat, 01 Nov 2025 00:00:00 GMT https://scholar.korea.ac.kr/handle/2021.sw.korea/281149 2025-11-01T00:00:00Z https://scholar.korea.ac.kr/handle/2021.sw.korea/270631 Title: Variable-selection consistency of linear quantile regression by validation set approach Authors: Kim, Suin; Lee, Sarang; Shin, Nari; Jung, Yoonsuh Abstract: We consider the problem of variable selection in the quantile regression model by cross-validation. Although cross-validation is commonly used in quantile regression for model selection, its theoretical justification has not yet been verified. In this work, we prove that cross-validation with the check loss function can lead to variable-selection consistency in quantile regression. Specifically, we investigate its asymptotic properties in linear quantile regression and its penalized version under both fixed and diverging number of parameters. For penalized models, penalties with the oracle property combined with cross-validation are shown to provide variable-selection consistency. In general, one of the crucial requirements for this consistency to hold is that the validation set size should be asymptotically equivalent to the total number of observations, which is also required in the conditional mean linear regression. Fri, 01 Aug 2025 00:00:00 GMT https://scholar.korea.ac.kr/handle/2021.sw.korea/270631 2025-08-01T00:00:00Z https://scholar.korea.ac.kr/handle/2021.sw.korea/274227 Title: 정책학 Authors: Haeil Jung Tue, 01 Jul 2025 00:00:00 GMT https://scholar.korea.ac.kr/handle/2021.sw.korea/274227 2025-07-01T00:00:00Z https://scholar.korea.ac.kr/handle/2021.sw.korea/269432 Title: Regional innovation and economic growth: Empirical insights from FGLS, FE-DKSE, and XGBoost-SHAP approach Authors: Pyo, Sangjae; Choi, Sang Ok Abstract: This study examines the impact of regional innovation factors and their dynamic interactions on regional economic growth. Using a 10-year panel dataset from 17 provinces and cities in South Korea, we analyze the data using panel data regression and explore non-linear relationships with the XGBoost model. We also enhance the model's interpretability by analyzing the impact of the variables on the predictions through Shapley Additive Explanations (SHAP) analysis. In the FGLS analysis, the number of innovative regional firms positively impacted GRDP, suggesting innovative firms boost regional economic output, but had minimal effect on Per Capita Gross Value Added (PCGVA). The proportion of top universities showed no significant impact on GRDP or PCGVA, though XGBoost and SHAP analyses indicated a complex, negative relationship with PCGVA. The interaction between the proportion of top universities and the number of innovative regional firms positively affected GRDP, emphasizing universities' role in supporting innovative firms. The effect of government R&D investment support on PCGVA highlights the importance of appropriate levels of support for maximizing productivity. The results indicate that R&D investment and open innovation significantly influence regional economic growth, with their effects exhibiting non-linear relationships. High-tech and medium-tech industries drive regional economic growth, while low-tech sectors have a negative impact on growth and per capita value-added. This study offers policy implications for fostering regional economic development as well as insights into the intricate links between economic growth and regional innovation factors. © 2025 Sun, 01 Jun 2025 00:00:00 GMT https://scholar.korea.ac.kr/handle/2021.sw.korea/269432 2025-06-01T00:00:00Z