• Title/Summary/Keyword: future information center

Search Result 1,243, Processing Time 0.03 seconds

Comparative Analysis of Multi-Agent Reinforcement Learning Algorithms Based on Q-Value (상태 행동 가치 기반 다중 에이전트 강화학습 알고리즘들의 비교 분석 실험)

  • Kim, Ju-Bong;Choi, Ho-Bin;Han, Youn-Hee
    • Proceedings of the Korea Information Processing Society Conference
    • /
    • 2021.05a
    • /
    • pp.447-450
    • /
    • 2021
  • 시뮬레이션을 비롯한 많은 다중 에이전트 환경에서는 중앙 집중 훈련 및 분산 수행(centralized training with decentralized execution; CTDE) 방식이 활용되고 있다. CTDE 방식 하에서 중앙 집중 훈련 및 분산 수행 환경에서의 다중 에이전트 학습을 위한 상태 행동 가치 기반(state-action value; Q-value) 다중 에이전트 알고리즘들에 대한 많은 연구가 이루어졌다. 이러한 알고리즘들은 Independent Q-learning (IQL)이라는 강력한 벤치 마크 알고리즘에서 파생되어 다중 에이전트의 공동의 상태 행동 가치의 분해(Decomposition) 문제에 대해 집중적으로 연구되었다. 본 논문에서는 앞선 연구들에 관한 알고리즘들에 대한 분석과 실용적이고 일반적인 도메인에서의 실험 분석을 통해 검증한다.

Confirmed cases of severe fever with thrombocytopenia syndrome in companion cats with a history of tick exposure in the Republic of Korea

  • Sun-Woo, Han;Ju-Hyun, An;Ji-Min, Rim;Eunseok, Jeong;Sungjun, Noh;Myoungdai, Kang;Jun-Gu, Kang;Joon-Seok, Chae
    • Journal of Veterinary Science
    • /
    • v.23 no.6
    • /
    • pp.83.1-83.7
    • /
    • 2022
  • Severe fever with thrombocytopenia syndrome (SFTS) is a zoonotic disease, and its clinical information and prevalence are important. This study was conducted on 22 feline patients from the Republic of Korea (ROK), suspected to suffer from a tick-borne disease. Four cats were positive for SFTS, and genotypes B-1, B-3, D, and F were identified. Clinical symptoms, such as anorexia, jaundice, thrombocytopenia, leukopenia, and hyperbilirubinemia, were detected. This is the first report of SFTS virus genotypes B-1, D, and F from cats in the ROK. Moreover, our results suggest that jaundice may be an indicator of SFTS in cats.

Economic Value Analysis of Supercomputing Service for Small and Medium-sized Businesses (중소기업 슈퍼컴퓨팅 서비스의 경제적 가치 분석)

  • Kim, Jae-Sung;Lee, Sang-Min;Kim, Myung-Il;Jang, Jin-Gyu
    • IE interfaces
    • /
    • v.23 no.4
    • /
    • pp.319-326
    • /
    • 2010
  • The SMB supercomputing service provides a supercomputing infrastructure for Small and Medium-sized Business (SMB) to enhance the efficiency of product development activity. In this study, the economic value of SMB supercomputing service is presented. The economic value of service is comprised of a direct and an indirect economic value. A direct economic value was estimated based on the result of user survey on service effectiveness. The input-output analysis method was adopted to estimate an indirect economic value induced from a direct economic value. The future economic value based on the result of user demand survey is also presented. As a result, the annual and future economic value of service were estimated about 55,215 million won and 323,928 million won, respectively.

The Effects of Information Quality on Call Center Agents' Individual Impacts : Focused on Call Center Customer Service Information Systems (콜센터 고객정보시스템의 정보품질이 상담원 업무 성과에 미치는 영향에 관한 연구)

  • Yang, Dong Hyun;An, Joon Mo;Hahm, Yu Kun;Min, Hyoung Jin
    • Journal of Information Technology Services
    • /
    • v.13 no.1
    • /
    • pp.87-101
    • /
    • 2014
  • This paper is a study on information Quality of Domestic Call Center Information System; recognized the importance of information quality and analyzed the impact on use and satisfaction of users following information quality which is a success element and ultimately the impact on user's performance by expanding information quality evaluating model by Jeong-Woo Lee et al.(2003) who reflected domestic special corporate culture based on information system success model by DeLone and McLean(2003). This is meaningful in providing theoretical foundation for efficient information quality supply and building quality management system at call center in future and provides hints to be considered in raising understanding of consultants on provided information and arranging ground that can utilize it in a timely manner.

Enabling Performance Intelligence for Application Adaptation in the Future Internet

  • Calyam, Prasad;Sridharan, Munkundan;Xu, Yingxiao;Zhu, Kunpeng;Berryman, Alex;Patali, Rohit;Venkataraman, Aishwarya
    • Journal of Communications and Networks
    • /
    • v.13 no.6
    • /
    • pp.591-601
    • /
    • 2011
  • Today's Internet which provides communication channels with best-effort end-to-end performance is rapidly evolving into an autonomic global computing platform. Achieving autonomicity in the Future Internet will require a performance architecture that (a) allows users to request and own 'slices' of geographically-distributed host and network resources, (b) measures and monitors end-to-end host and network status, (c) enables analysis of the measurements within expert systems, and (d) provides performance intelligence in a timely manner for application adaptations to improve performance and scalability. We describe the requirements and design of one such "Future Internet performance architecture" (FIPA), and present our reference implementation of FIPA called 'OnTimeMeasure.' OnTimeMeasure comprises of several measurement-related services that can interact with each other and with existing measurement frameworks to enable performance intelligence. We also explain our OnTimeMeasure deployment in the global environment for network innovations (GENI) infrastructure collaborative research initiative to build a sliceable Future Internet. Further, we present an applicationad-aptation case study in GENI that uses OnTimeMeasure-enabled performance intelligence in the context of dynamic resource allocation within thin-client based virtual desktop clouds. We show how a virtual desktop cloud provider in the Future Internet can use the performance intelligence to increase cloud scalability, while simultaneously delivering satisfactory user quality-of-experience.

Research on PEFT Feasibility for On-Device Military AI (온 디바이스 국방 AI를 위한 PEFT 효용성 연구)

  • Gi-Min Bae;Hak-Jin Lee;Sei-Ok Kim;Jang-Hyong Lee
    • Proceedings of the Korean Society of Computer Information Conference
    • /
    • 2024.01a
    • /
    • pp.51-54
    • /
    • 2024
  • 본 논문에서는 온 디바이스 국방 AI를 위한 효율적인 학습 방법을 제안한다. 제안하는 방법은 모델 전체를 재학습하는 대신 필요한 부분만 세밀하게 조정하여 계산 비용과 시간을 대폭 줄이는 PEFT 기법의 LoRa를 적용하였다. LoRa는 기존의 신경망 가중치를 직접 수정하지 않고 추가적인 낮은 랭크의 매트릭스를 학습하는 방식으로 기존 모델의 구조를 크게 변경하지 않으면서도, 효율적으로 새로운 작업에 적응할 수 있다. 또한 학습 파라미터 및 연산 입출력에 데이터에 대하여 32비트의 부동소수점(FP32) 대신 부동소수점(FP16, FP8) 또는 정수형(INT8)을 활용하는 경량화 기법인 양자화도 적용하였다. 적용 결과 학습시 요구되는 GPU의 사용량이 32GB에서 5.7GB로 82.19% 감소함을 확인하였다. 동일한 조건에서 동일한 데이터로 모델의 성능을 평가한 결과 동일 학습 횟수에선 LoRa와 양자화가 적용된 모델의 오류가 기본 모델보다 53.34% 증가함을 확인하였다. 모델 성능의 감소를 줄이기 위해서는 학습 횟수를 더 증가시킨 결과 오류 증가율이 29.29%로 동일 학습 횟수보다 더 줄어듬을 확인하였다.

  • PDF

A Study of 4G Network for Security System

  • Kim, Suk-jin;Lee, Hyangran;Lee, Malrey
    • International Journal of Advanced Culture Technology
    • /
    • v.3 no.2
    • /
    • pp.77-86
    • /
    • 2015
  • In this paper there is an overview of some standards and security models which are implemented in such an IP-based and heterogeneous networks and we also present some security models in an open environment and finally we obtain that as a result of the nature of 4G networks there are still more security holes and open issues for expert to notice. Our survey shows that a number of new security threats to cause unexpected service interruption and disclosure of information will be possible in 4G due mainly to the fact that 4G is an IP-based, heterogeneous network. Other than that, it tells about the security issues and vulnerabilities present in the above 4G standards are discussed. Finally, we point to potential areas for future vulnerabilities and evaluate areas in 4G security which warrant attention and future work by the research and advanced technology industry.

A Study on the Revision of UDC Korean Edition (UDC 한국어판의 개정에 관한 연구)

  • Lee, Chang-Soo
    • Journal of Information Management
    • /
    • v.41 no.3
    • /
    • pp.1-26
    • /
    • 2010
  • The purpose of this study is to compare and analyze the print form of UDC(Universal Decimal Classification) standard edition which was published by British Standards Institution in 2005 to Korean edition which was published by Korea Scientific & Technological Information Center in 1973, and to suggest the appropriate revision directions of future Korean edition. This study suggests that the future Korean edition should be revised based on the Master Reference File and should be a print edition which is composed of systematic tables and alphabetical index from the standard edition. In addition, the future Korean edition needs to strengthen international universality and to extend synthetic method using its auxiliary tables.

Uncertainty Characteristics in Future Prediction of Agrometeorological Indicators using a Climatic Water Budget Approach (기후학적 물수지를 적용한 기후변화에 따른 농업기상지표 변동예측의 불확실성)

  • Nam, Won-Ho;Hong, Eun-Mi;Choi, Jin-Yong;Cho, Jaepil;Hayes, Michael J.
    • Journal of The Korean Society of Agricultural Engineers
    • /
    • v.57 no.2
    • /
    • pp.1-13
    • /
    • 2015
  • The Coupled Model Intercomparison Project Phase 5 (CMIP5), coordinated by the World Climate Research Programme in support of the Intergovernmental Panel on Climate Change (IPCC) AR5, is the most recent, provides projections of future climate change using various global climate models under four major greenhouse gas emission scenarios. There is a wide selection of climate models available to provide projections of future climate change. These provide for a wide range of possible outcomes when trying to inform managers about possible climate changes. Hence, future agrometeorological indicators estimation will be much impacted by which global climate model and climate change scenarios are used. Decision makers are increasingly expected to use climate information, but the uncertainties associated with global climate models pose substantial hurdles for agricultural resources planning. Although it is the most reasonable that quantifying of the future uncertainty using climate change scenarios, preliminary analysis using reasonable factors for selecting a subset for decision making are needed. In order to narrow the projections to a handful of models that could be used in a climate change impact study, we could provide effective information for selecting climate model and scenarios for climate change impact assessment using maximum/minimum temperature, precipitation, reference evapotranspiration, and moisture index of nine Representative Concentration Pathways (RCP) scenarios.