• Journal of KIISE:Software and Applications
• /
• v.35 no.3
• /
• pp.179-188
• /
• 2008

An important issue in multiagent reinforcement learning is how an agent should team its optimal policy through trial-and-error interactions in a dynamic environment where there exist other agents able to influence its own performance. Most previous works for multiagent reinforcement teaming tend to apply single-agent reinforcement learning techniques without any extensions or are based upon some unrealistic assumptions even though they build and use explicit models of other agents. In this paper, basic concepts that constitute the common foundation of multiagent reinforcement learning techniques are first formulated, and then, based on these concepts, previous works are compared in terms of characteristics and limitations. After that, a policy model of the opponent agent and a new multiagent reinforcement learning method using this model are introduced. Unlike previous works, the proposed multiagent reinforcement learning method utilize a policy model instead of the Q function model of the opponent agent. Moreover, this learning method can improve learning efficiency by using a simpler one than other richer but time-consuming policy models such as Finite State Machines(FSM) and Markov chains. In this paper. the Cat and Mouse game is introduced as an adversarial multiagent environment. And effectiveness of the proposed multiagent reinforcement learning method is analyzed through experiments using this game as testbed.

• Korean Journal of General Education
• /
• v.4 no.1
• /
• pp.237-254
• /
• 2010

For years, poor research and working conditions in the field of science and engineering in Korea led to the migration of gifted students to seek a better academic environment. The primary reason for such a phenomenon is the lack of a coherent education system for gifted students. With the support from the Ministry of Education, Science and Technology and the Korea Foundation for the Advancement of Science & Creativity, Hanyang University launched an exploring project to introduce an Honors Program, aiming to provide specialized and systematic learning opportunities as well as supporting greater academic achievements for gifted students at university-level. Students admitted to this program were selected not by conventional standards but by competency-based selection process, assessed through an in-depth interview. The specific goal of this program is to prepare gifted and talented students not only for an academic pioneer with the highest caliber in the field of science, but also for a global leader equipped with a creative view as well as integrity and a convergent mind. Distinctive features of this program include emphasis on fundamental science and consilience, problem solving skills, first-hand education, interpersonal skills, and global communication skills as well as individualization of the learning experience, among many others. This paper provides a short glimpse of the focus and methodology Honors Program in Hanyang University offers.

• The Journal of the Korea Contents Association
• /
• v.13 no.8
• /
• pp.474-481
• /
• 2013

The purpose of this study is that pre-service teacher are able to design creative instruction based on NFTM-TRIZ for building up their continuous creative thinking and promoting their creative instruction activities. NFTM-TRIZ is a educational technology system to form and develop creative thinking from child to adult continuously based on TRIZ theory. TRIZ is the thinking technique of creative problem solving that can be the tool of inventory solutions by finding and get over the key of contradiction that is necessary to obtain ideal final results of suggested problems. The subjects for this study were 90 pre-service teachers who are attending third and fourth graders of Teachers' College in G university and are taking 'Curriculum and Educational Evaluation'. The creativity program for this study was carried out for ten minutes at the end of lectures. The verification for this study results were performed two faces. First, pre-service teachers presented teaching and learning plan for one time used 8 Steps' Teaching and Learning Model based on NFTM-TRIZ. Second, researcher got feedback from them about this creative program.

• Journal of KIISE:Software and Applications
• /
• v.30 no.3_4
• /
• pp.316-325
• /
• 2003

A user preference prediction method using an exiting collaborative filtering technique has used the nearest-neighborhood method based on the user preference about items and has sought the user's similarity from the Pearson correlation coefficient. Therefore, it does not reflect any contents about items and also solve the problem of the sparsity. This study suggests the preference prediction system using the similarity weight granted Bayesian estimated value and the associative user clustering to complement problems of an exiting collaborative preference prediction method. This method suggested in this paper groups the user according to the Genre by using Association Rule Hypergraph Partitioning Algorithm and the new user is classified into one of these Genres by Naive Bayes classifier to slove the problem of sparsity in the collaborative filtering system. Besides, for get the similarity between users belonged to the classified genre and new users, this study allows the different estimated value to item which user vote through Naive Bayes learning. If the preference with estimated value is applied to the exiting Pearson correlation coefficient, it is able to promote the precision of the prediction by reducing the error of the prediction because of missing value. To estimate the performance of suggested method, the suggested method is compared with existing collaborative filtering techniques. As a result, the proposed method is efficient for improving the accuracy of prediction through solving problems of existing collaborative filtering techniques.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.2
• /
• pp.631-638
• /
• 2017

In the age of the Fourth Industrial Revolution, lifestyle and industrial structures are faced with evolution from IT-based automation to the intelligent stage, demanding talents with software capabilities in various fields. Reflecting these demands, the government has enhanced basic software education for non-majors in elementary and secondary schools as well as universities. In this study, the software convergence education of Non-Majors is proposed to improve the general problem solving ability based on computational thinking and the software convergence ability in the field of their own by developing robot activity. The subjects of this study were 91 students, who were composed of various majors. The class was designed with computing thinking, convergence elements, and creative robot activity. The study was conducted for 13 weeks. To examine the effects of software convergence education through the creative robot activity, this study observed changes in the students' learning outcomes, satisfaction with creative robot activities, and perceptions of other disciplines after class based on pre-diagnosis surveys. The survey asked 12 questions including an understanding of the learning contents, overall satisfaction with multidisciplinary collaborative learning, understanding of other disciplines, and self-evaluation of problem solving ability through creative robot activities, which were compared with that before the class. They answered that their ability was improved.

• The Journal of Korean Association of Computer Education
• /
• v.11 no.1
• /
• pp.57-64
• /
• 2008

This paper presents a web-based virtual educational system for Java language, which consists of a management system named Java Web Player (JWP) and creative multimedia contents for the lectures of Java language. The JWP is a Java application program free from security problems by the Java Web Start technologies that supports an integrated learning environment including three important learning procedures: Java concept learning process, programming practice process and assessment process. On-line voice presentation and its related texts together with moving images are synchronized for efficiently conveying creative contents to learners. Furthermore, a simple and useful compiler is included in the JWP for providing user-friendly language practice environment enabling such as coding, editing, executing, and debugging Java source files on the Web. Finally, simple multiple choices are given suddenly to the learners while they are studying through the JWP and the test results are displayed on the message box. In order to show the validity of the proposed virtual educational system we analysed and assessed the learners' academic performance on the five quizzes for one semester.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.181-185
• /
• 2016

최근 교육과정에서 학생들의 능력 평가는 단순 암기보다 학생들의 종합적인 사고력을 판단할 수 있는 서답형 문항을 늘리는 방향으로 변하고 있다. 그러나 서답형 문항의 경우 채점하는 데 시간과 비용이 많이 들고, 채점자의 주관에 따라 채점 결과의 일관성과 신뢰성을 보장하기 어렵다는 문제가 있다. 이런 점을 해결하기 위해 해외의 사례를 참고하여 국내에서도 서답형 문항에 자동채점 시스템을 적용하는 연구를 진행하고 있다. 본 논문에서는 2014년도에 개발된 '한국어 문장 수준 서답형 문항 자동채점 시스템'의 성능분석을 바탕으로 언어 처리 기능과 자동채점 성능을 개선한 2015년도 자동채점 시스템을 간략하게 소개하고, 각 자동채점 시스템의 성능을 비교 분석한다. 성능 분석 대상으로는 2014년도 국가수준 학업성취도평가의 서답형 문항을 사용했다. 실험 결과, 개선한 시스템의 평균 완전 일치도와 평균 정확률이 기존의 시스템보다 각각 9.4%p, 8.9%p 증가했다. 자동채점 시스템의 목적은 가능한 채점 시간을 단축하면서 채점 기준의 일관성과 신뢰성을 확보하는 데 있으므로, 보완한 2015년 자동채점 시스템의 성능이 향상되었다고 판단할 수 있다.

• Journal of Convergence for Information Technology
• /
• v.11 no.9
• /
• pp.28-34
• /
• 2021

In the case of object detection using deep learning, both accuracy and real-time are required. However, it is difficult to use a deep learning model that processes a large amount of data in a limited resource environment. To solve this problem, this paper proposes an object detection model for small embedded devices. Unlike the general detection model, the model size was minimized by using a structure in which the pre-trained feature extractor was removed. The structure of the model was designed by repeatedly stacking lightweight convolution blocks. In addition, the number of region proposals is greatly reduced to reduce detection overhead. The proposed model was trained and evaluated using the public dataset PASCAL VOC. For quantitative evaluation of the model, detection performance was measured with average precision used in the detection field. And the detection speed was measured in a Raspberry Pi similar to an actual embedded device. Through the experiment, we achieved improved accuracy and faster reasoning speed compared to the existing detection method.

• Korean Journal of Remote Sensing
• /
• v.39 no.5_4
• /
• pp.1111-1123
• /
• 2023

In recent times, semantic image segmentation methods using deep learning models have been widely used for monitoring changes in surface attributes using remote sensing imagery. To enhance the performance of various UNet-based deep learning models, including the prominent UNet model, it is imperative to have a sufficiently large training dataset. However, enlarging the training dataset not only escalates the hardware requirements for processing but also significantly increases the time required for training. To address these issues, transfer learning is used as an effective approach, enabling performance improvement of models even in the absence of massive training datasets. In this paper we present three transfer learning models, UNet-ResNet50, UNet-VGG19, and CBAM-DRUNet-VGG19, which are combined with the representative pretrained models of VGG19 model and ResNet50 model. We applied these models to building extraction tasks and analyzed the accuracy improvements resulting from the application of transfer learning. Considering the substantial impact of learning rate on the performance of deep learning models, we also analyzed performance variations of each model based on different learning rate settings. We employed three datasets, namely Kompsat-3A dataset, WHU dataset, and INRIA dataset for evaluating the performance of building extraction results. The average accuracy improvements for the three dataset types, in comparison to the UNet model, were 5.1% for the UNet-ResNet50 model, while both UNet-VGG19 and CBAM-DRUNet-VGG19 models achieved a 7.2% improvement.

• Journal of the Korean Society for Library and Information Science
• /
• v.56 no.2
• /
• pp.5-33
• /
• 2022

In this study, the direction and contents of public education to develop competencies that are in line with the 2022 revised curriculum and the paradigm of future education were sought, and a plan for reading education was prepared. A creative and cooperative problem-solving method and process as not only information literacy ability to read, select, and reconstruct information but also transformative competency to respond to uncertain future and diversified situations are important amid the complex, pluralistic and rapid development of information and communication technology It was intended to give an experience of exploring and communicating through reading in order to explore and derive it. Analyze the general outline and syllabus of the curriculum, international education project definitions and indicators, and organize academic theories and research to set the direction and goal of high school reading education, and organize creative and convergence class strategies and reading activities A library reading class model was developed. Accordingly, the class model was revised by applying the development research method, and the final model was developed by supplementing it through field application evaluation. In order to achieve the research purpose end, a two-round Delphi survey was conducted on 10 reading education and curriculum experts. The model modified through the Delphi survey was developed in the final school library convergence reading class model by demonstrating the class in the educational field and supplementing it through application evaluation.