• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2002.05a
• /
• pp.6-9
• /
• 2002

오랜 세월동안 위대한 이동수단을 만들어내고자 하는 인간의 끓은 오늘날 눈부신 각종 운송기구를 만들어 내는 결실을 얻고 있다. 자동차 네비게이션 시스템도 그러한 결실중의 한 예라고 할 수 있을 것이다. 지능적으로 판단하고 정보를 처리할 수 있는 자동차 네비게이션 시스템을 부착함으로써 한단계 발전한 운송수단으로 진화할 수 있을 것이다. 이러한 자동차 네비게이션 시스템의 단점이라면 한정된 리 소스만으로 여러 가지 작업을 수행해야만 하는 어려움이다. 그래서 네비게이션 시스템의 주요 작업중의 하나인 경로를 추출하는 경로추출(Route Planing) 작업은 한정된 리 소스에서도 최적의 경로를 찾을 수 있는 지능적인 방법이어야만 한다. 이러한 경로를 추출하는 작업을 하는 데 기존에 일반적으로 쓰였던 두 가지 방법에는 Dijkstra's algorithm과 A* algorithm이 있다. 이 두 방법은 최적의 경로를 찾아 낸다는 점은 있지만 경로를 찾기 위해서 알고리즘의 특성상 각각, 넓은 영역에 대하여 탐색작업을 해야하고 또한 수행시간이 많이 걸린다는 단점과 또한 경로를 계산하기 위해서 Heuristic function을 추가적인 정보로 계산을 해야 한다는 단점이 있다. 본 논문에서는 적은 탐색 영역을 가지면서 또한 최적의 경로를 추출하는 데 드는 수행시간은 작으며 나아가 동적인 교통환경에서도 최적의 경로를 추출할 수 있는 최적 경로 추출방법을 강화학습의 일종인 Q- Learning을 이용하여 구현해 보고자 한다.

• Journal of Korean Elementary Science Education
• /
• v.25 no.spc5
• /
• pp.548-555
• /
• 2007

In this paper, we suggested that those elementary school students who are gifted in science should be taught basic and fundamental concepts to solve applied problems. We developed a teaching model based on a lesson regarding the path that light takes when passing through a lens on the base of refraction of light. We applied the teaching model to scientifically-gifted elementary school students and analyzed the results. The teaching model is based on the circulation loaming model appropriate for learning such concepts. The problems were designed and applied in order to determine the students' level of concept skills held and also to develop new teaching tools to help their understanding of concepts. As a result, we confirmed that the students, who were unable to describe the path of the light before the course of instruction was given, were able to draw and explain the path of light passing trough lens by using the law of refraction following the instruction.

• Journal of Korean Library and Information Science Society
• /
• v.45 no.1
• /
• pp.5-25
• /
• 2014

The goal of this study is to propose the plan for activating collection services of public library in Seoul. For this goal, author analyzed and compared the core infrastructure, collection services, and their correlation of public libraries in 16 local governments and evaluated environment and the current situation of collection service of public libraries in Seoul. Based on the these results, author suggested five plans or strategies for vitalizing the collection services of public libraries. All public libraries in Seoul must maximize the visibility of the new collections, supply the culture(life-long learning) programs based on library collections, expand the breadth of the collection service through library cooperation system, strengthen interlibrary loan by establishment of the collaborative repository, and raise the awareness of the citizens about the library and collections.

• Proceedings of the Korean Information Science Society Conference
• /
• 2007.06c
• /
• pp.319-323
• /
• 2007

실세계의 여러 문제들은 마르코프 결정 문제(Markov decision problem, MDP)로 표현될 수 있고, 이 MDP는 모델이 알려진 경우에는 평가치 반복(value iteration) 이나 모델이 알려지지 않은 경우에도 강화 학습(reinforcement learning) 알고리즘 등을 사용하여 풀 수 있다. 하지만 이들 알고리즘들은 시간 복잡도가 높아 크기가 큰 실세계 문제에 적용하기 쉽지 않아, MDP를 계층적으로 분할하거나, 여러 단계를 묶어서 수행하는 등의 시간적 추상화(temporal abstraction) 방법이 제안되어 왔다. 이러한 시간적 추상화 방법들의 문제점으로는 시간적 추상화의 디자인에 따라 MDP의 풀이 성능이 크게 달라질 수 있으며, 많은 경우 사용자가 이 디자인을 직접 제공해야 한다는 것들이 있다. 최근 사용자의 간섭이 필요 없이 자동적으로 시간적 추상화를 만드는 방법들이 제안된 바 있으나, 이들 방법들 역시 결과물에 대한 이론적인 성능 보장(performance guarantee)은 제공하지 못하고 있다. 본 연구에서는 이러한 문제점을 해결하기 위해 MDP의 구조와 그 풀이 성능을 연관짓는 복잡도 척도에 대해 살펴본다. 이를 위해 MDP로부터 얻은 상태 경로 그래프(state trajectory graph)의 위상적 성질들을 여러 네트워크 척도(network measurements) 들을 이용하여 측정하고, 이와 MDP의 풀이 성능과의 관계를 다양한 상황에 대해 실험적, 이론적으로 분석해 보았다.

• The Journal of Korea Robotics Society
• /
• v.15 no.2
• /
• pp.197-204
• /
• 2020

In this paper, we present a learning platform for robotic grasping in real world, in which actor-critic deep reinforcement learning is employed to directly learn the grasping skill from raw image pixels and rarely observed rewards. This is a challenging task because existing algorithms based on deep reinforcement learning require an extensive number of training data or massive computational cost so that they cannot be affordable in real world settings. To address this problems, the proposed learning platform basically consists of two training phases; a learning phase in simulator and subsequent learning in real world. Here, main processing blocks in the platform are extraction of latent vector based on state representation learning and disentanglement of a raw image, generation of adapted synthetic image using generative adversarial networks, and object detection and arm segmentation for the disentanglement. We demonstrate the effectiveness of this approach in a real environment.

• The Journal of the Korea Contents Association
• /
• v.20 no.5
• /
• pp.236-242
• /
• 2020

Big data technology has been rising as a leading technology in the healthcare paradigm. As a world-class big data nation including National Health Insurance data, Korea has been focused on health policies and sustainability through database forecasting and policy establishment. So the need for education of big data by public officials in healthcare sector is increasing. However, there has not yet been National Competency Standards(NCS) or education modules, in this study, healthcare big data education module and content have been developed for the public servants with confidence.

• Journal of Korean Library and Information Science Society
• /
• v.32 no.4
• /
• pp.303-329
• /
• 2001

The purpose of this study was to apply internet technology to school library, and to construct the contents of KISLMC(Korean Internet School Library Media Center). The content of KISLMC was composed of three categories, Cyber library, Supporting the materials for teaching and learning, and Guidance and counseling to the students. To develop the user-oriented interface, interface metaphors such as icons and images were used. Also, sitemap, search engine, index were used to avoid to be lost during the navigation.

• Journal of Korea Multimedia Society
• /
• v.21 no.8
• /
• pp.969-981
• /
• 2018

Recent advances in machine learning approaches such as deep neural network and reinforcement learning offer significant performance improvements in generating detailed and varied motions in physically simulated virtual environments. The optimization methods are highly attractive because it allows for less understanding of underlying physics or mechanisms even for high-dimensional subtle control problems. In this paper, we propose an efficient learning method for stochastic policy represented as deep neural networks so that agent can generate various energetic motions adaptively to the changes of tasks and states without losing interactivity and robustness. This strategy could be realized by our novel trajectory search method motivated by the trust region policy optimization method. Our value-based trajectory smoothing technique finds stably learnable trajectories without consulting neural network responses directly. This policy is set as a trust region of the artificial neural network, so that it can learn the desired motion quickly.

• Journal of the Korean BIBLIA Society for library and Information Science
• /
• v.15 no.2
• /
• pp.255-270
• /
• 2004

American Academic Accreditation System was reviewed. Current accreditation standards of the six regional accreditation associations were analysed with reference to the evaluation of academic libraries. That analysis was led to the recognition of some trends; mission and goal based assessment, qualitative evaluation, emphasis on the student learning outcomes, emphasis on the teaching role of academic librarians, distance education, and less prescriptive text. Some suggestions were made for revision of the Korean academic accreditation standards related to libraries.

• Journal of The Korean Association of Information Education
• /
• v.11 no.2
• /
• pp.243-249
• /
• 2007

Most homepages of schools are available to publicize activities of education, to load educational materials efficiently. But we can't use current homepages of schools in making characteristic information or sharing educational resources. In this study, this I-CUBE system with RSS was designed and embodied for sharing characteristic educational activities and works in schools. Currently, I-CUBE is being under experimental test for sharing education activities and making community in elementry schools,.