• Title/Summary/Keyword: 융합 전공

Search Result 1,624, Processing Time 0.028 seconds

A Study on Learning Performance Improvement by Using Hidden States in Deep Reinforcement Learning (심층강화학습에 은닉 상태 정보 활용을 통한 학습 성능 개선에 대한 고찰)

  • Choi, Yohan;Seok, Yeong-Jun;Kim, Ju-Bong;Han, Youn-Hee
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2022.05a
    • /
    • pp.528-530
    • /
    • 2022
  • 심층강화학습에 완전 연결 신경망과 합성곱 신경망은 잘 활용되는 것에 반해 순환 신경망은 잘 활용되지 않는다. 이는 강화학습이 마르코프 속성을 전제로 하기 때문이다. 지금까지의 강화학습은 환경이 마르코프 속성을 만족하도록 사전 작업이 필요했다, 본 논문에서는 마르코프 속성을 따르지 않는 환경에서 이러한 사전 작업 없이도 순환 신경망의 은닉 상태를 통해 마르코프 속성을 학습함으로써 학습 성능을 개선할 수 있다는 것을 소개한다.

Analysis of Domestic and Overseas Disaster Research Trends - Focusing on ICT (국내·외 재난 연구 동향 분석 - ICT 중심으로)

  • Kim, Gwan-Jun;Lee, Gang-Jun;Back, Sung-Won;Sim, Young-Mi;Jeong, Sang
    • Proceedings of the Korean Society of Disaster Information Conference
    • /
    • 2023.11a
    • /
    • pp.195-197
    • /
    • 2023
  • 본 연구는 재난에서 사용되고 있는 정보통신기술( ICT) 연구 국내논문 100개 및 해외논문 100개를 수집한 후, R 프로그램을 이용한 텍스트 마이닝 분석을 통해 국내 재난관리 네 가지 단계에 따른 ICT 기술 관련 연구 동향을 파악하고, 해외 재난관리의 ICT 기술 연구 동향을 분석 활용하여 우리나라 재난에서의 ICT 기술 활용방안을 새롭게 구성해 제언하는 것이다.

  • PDF

Design of Clustering CoaT Vision Model Based on Transformer (Transformer 기반의 Clustering CoaT 모델 설계)

  • Bang, Ji-Hyeon;Park, Jun;Jung, Se-Hoon;Sim, Chun-Bo
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2022.05a
    • /
    • pp.546-548
    • /
    • 2022
  • 최근 컴퓨터 비전 분야에서 Transformer를 도입한 연구가 활발히 연구되고 있다. 이 모델들은 Transformer의 구조를 거의 그대로 사용하기 때문에 확장성이 좋으며 large 스케일 학습에서 매우 우수한 성능을 보여주었다. 하지만 Transformer를 적용한 비전 모델은 inductive bias의 부족으로 학습 시 많은 데이터와 시간을 필요로 하였다. 그로 인하여 현재 많은 Vision Transformer 개선 모델들이 연구되고 있다. 본 논문에서도 Vision Transformer의 문제점을 개선한 Clustering CoaT 모델을 제안한다.

Convergence Study on Major Satisfaction and Academic Achievement Depending on the Characteristics of Community Service Experience in University students (대학생의 사회봉사활동 경험특성에 따른 전공만족도 및 학업성취도에 관한 융합연구)

  • Heo, Seong-Eun
    • Journal of the Korea Convergence Society
    • /
    • v.7 no.4
    • /
    • pp.85-96
    • /
    • 2016
  • This study aimed to provide basic data required for operation and development of a program that is necessary for linkage with the major as well as proper settlement of community service of the university by identifying university students' major satisfaction and academic achievement level in relation with the characteristics of community service. A survey was conducted in 401 university students who voluntarily participated in this study using the online survey panel targeting university students who reside in Busan and South Gyeongsang Province from December 1, 2014 to February 1, 2015. The results are as follows. Students who had a community service experience and had higher participation in community service showed higher major satisfaction and academic achievement. Particularly, when community service was related to the major, major satisfaction was higher. In addition, the higher the satisfaction with community service was and the higher the need of community service was, the higher the major satisfaction was. Students with health major showed higher major satisfaction than students with non-health major. Therefore, development of an effective program will be necessary to increase the need and participation of community service and revitalize community service more continuously.

Risk Assessment of Heavy Metals Migrated from Plastic Food Utensils, Containers, and Packaging Distributed in Korea (국내 유통 식품용 플라스틱 기구 및 용기, 포장의 중금속 위해도 평가)

  • Kyung Youn, Lee;Hyung Soo, Kim;Dae Yong, Jang;Ye Ji, Koo;Seung Ha, Lee;Hye Bin, Yeo;Ji Su, Yoon;Kyung-Min, Lim;Jaeyun, Choi
    • KOREAN JOURNAL OF PACKAGING SCIENCE & TECHNOLOGY
    • /
    • v.28 no.3
    • /
    • pp.175-182
    • /
    • 2022
  • Heavy metals can be intentionally or unintentionally introduced into plastic food utensils, containers, and packaging (PFUCP) as additives or contaminants, which can be ingested with food by humans. Here, seven-heavy metals (lead, cadmium, nickel, chromium, antimony, copper, and manganese) with toxicity concerns were selected, and risk assessment was done by establishing their migration from 137 PFUCP products made of 16 materials distributed in Korea. Migration of heavy metals was examined by applying 4% acetic acid as a food simulant (70℃, 30 minutes) to the PFUCP products. Inductively coupled plasma mass spectrometry (ICP-MS) was employed for the analysis of migrated heavy metals, and the reliability of quantitative results was confirmed by checking linearity, LOD, LOQ, recovery, precision, and expanded uncertainty. As a result of monitoring, heavy metals were detected at a level of non-detection to 8.76 ± 11.87 ㎍/L and most of the heavy metals investigated were only detected at trace amounts of less than 1 ㎍/L on average. However, antimony migrated from PET products was significantly higher than other groups. Risk assessment revealed that all the heavy metals investigated were safe with a margin of exposure above 311. Collectively, we demonstrated that heavy metals migrated from PFUCP products distributed in Korea appear to be within the safe range.

A Study on the Awareness of Material Safety Data Sheet (MSDS) of University Laboratory Workers (대학 연구활동종사자의 물질안전보건자료(MSDS) 인식에 관한 연구)

  • Kim, Hong-Kwan;Chon, Young-Woo;Ko, Kwang-Hoon;Hwang, Yong-Woo;Kim, Jung-Soo;Lee, Ik-Mo
    • Korean Journal of Hazardous Materials
    • /
    • v.6 no.2
    • /
    • pp.87-94
    • /
    • 2018
  • The study is aimed to analyze the MSDS awareness as per the properties, including general characteristics, of laboratory workers in university. It conducted the self-administered survey on the subject of 780 laboratory workers engaging in research activities in three university from February to May, 2017. Among them, it analyzed 686 cases except 94 of them, lacking content. As a result, education experience of MSDS, necessity of education, usage experience of MSDS, installation, comprehension, and accessibility showed a strategically significant difference in the recognition of MSDS. According to the categorical importance of MSDS, "the first aid measures" directly related to life was most significantly considered. It is estimated that it will contribute to prevention of safety accident by strengthening education on MSDS and increasing the level of awareness for laboratory workers.

The Impact of Failure Frequency Items on Availability and Operation Support Costs of Armored Vehicles (장갑차의 가용도와 운영유지비용에 미치는 고장 다빈도 품목의 영향성 분석)

  • Bong, Ju-Sung;Baek, Il-Ho;Kim, Min-Seop;Hur, Jang-Wook
    • Journal of the Korean Society of Manufacturing Process Engineers
    • /
    • v.20 no.4
    • /
    • pp.8-15
    • /
    • 2021
  • The effects on system availability, operation, and support costs were analyzed using the M&S system (MPS). The failure frequency items of current armored vehicles were identified and the MTBF of the identified items was improved. The results of this study suggest that when we reduce the frequency of failure, the availability increases, and the operation and support costs decrease. By improving the reliability of the failure frequency items, it becomes possible to upgrade or develop the weapons systems. Through this study, we confirmed that improving reliability will enhance combat readiness and reduce operation and support costs.

A Comparative Study of Dietary Related Zero-waste Patterns and Consumer Responses Before and After COVID-19 (코로나-19 이전과 이후 식생활 관련 제로웨이스트 운동 양상과 소비자 반응 비교)

  • Park, In-Hyoung;Park, You-min;Lee, Cheol;Sun, Jung-eun;Hu, Wendie;Chung, Jae-Eun
    • Human Ecology Research
    • /
    • v.60 no.1
    • /
    • pp.21-38
    • /
    • 2022
  • This study uses text mining compares and contrasts consumers' social media discourses on dietary related zero-waste movement before and after COVID-19. The results indicate that the amount of buzz on social networks for the zero- waste movement has been increasing after COVID-19. Additionally, the results of frequency analysis and topic modeling revealed that subjects associated with zero-waste movement were more diversified after COVID-19. Although the results of a sentiment analysis and word cloud visualization confirmed that consumers' positive responses toward the zero-waste have been increasing, they also revealed a need to educate and encourage those who are still not aware of the need for zero-waste. Finally, consumers mentioned only a small number of companies participating in zero-waste movement on SNS, indicating that the level of active involvement by such companies is much lower than that of consumers. Theoretical and educational implications as well as those for government policy-making are considered.

Data Augmentation Scheme for Semi-Supervised Video Object Segmentation (준지도 비디오 객체 분할 기술을 위한 데이터 증강 기법)

  • Kim, Hojin;Kim, Dongheyon;Kim, Jeonghoon;Im, Sunghoon
    • Journal of Broadcast Engineering
    • /
    • v.27 no.1
    • /
    • pp.13-19
    • /
    • 2022
  • Video Object Segmentation (VOS) task requires an amount of labeled sequence data, which limits the performance of the current VOS methods trained with public datasets. In this paper, we propose two effective data augmentation schemes for VOS. The first augmentation method is to swap the background segment to the background from another image, and the other method is to play the sequence in reverse. The two augmentation schemes for VOS enable the current VOS methods to robustly predict the segmentation labels and improve the performance of VOS.

Super Resolution Performance Analysis of GAN according to Feature Extractor (특징 추출기에 따른 SRGAN의 초해상 성능 분석)

  • Park, Sung-Wook;Kim, Jun-Yeong;Park, Jun;Jung, Se-Hoon;Sim, Chun-Bo
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2022.11a
    • /
    • pp.501-503
    • /
    • 2022
  • 초해상이란 해상도가 낮은 영상을 해상도가 높은 영상으로 합성하는 기술이다. 딥러닝은 영상의 해상도를 높이는 초해상 기술에도 응용되며 실현은 2아4년에 발표된 SRCNN(Super Resolution Convolutional Neural Network) 모델로부터 시작됐다. 이후 오토인코더 (Autoencoders) 구조로는 SRCAE(Super Resolution Convolutional Autoencoders), 합성된 영상을 실제 영상과 통계적으로 구분되지 않도록 강제하는 GAN (Generative Adversarial Networks) 구조로는 SRGAN(Super Resolution Generative Adversarial Networks) 모델이 발표됐다. 모두 SRCNN의 성능을 웃도는 모델들이나 그중 가장 높은 성능을 끌어내는 SRGAN 조차 아직 완벽한 성능을 내진 못한다. 본 논문에서는 SRGAN의 성능을 개선하기 위해 사전 훈련된 특징 추출기(Pre-trained Feature Extractor) VGG(Visual Geometry Group)-19 모델을 변경하고, 기존 모델과 성능을 비교한다. 실험 결과, VGG-19 모델보다 윤곽이 뚜렷하고, 실제 영상과 더 가까운 영상을 합성할 수 있는 모델을 발견할 수 있을 것으로 기대된다.