• ETRI Journal
• /
• v.45 no.5
• /
• pp.735-745
• /
• 2023

This paper surveys recent multiagent reinforcement learning and neural Myerson auction deep learning efforts to improve mobility control and resource management in autonomous ground and aerial vehicles. The multiagent reinforcement learning communication network (CommNet) was introduced to enable multiple agents to perform actions in a distributed manner to achieve shared goals by training all agents' states and actions in a single neural network. Additionally, the Myerson auction method guarantees trustworthiness among multiple agents to optimize rewards in highly dynamic systems. Our findings suggest that the integration of MARL CommNet and Myerson techniques is very much needed for improved efficiency and trustworthiness.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.27 no.4
• /
• pp.474-484
• /
• 2024

Reinforcement learning, which are also studied in the field of defense, face the problem of sample efficiency, which requires a large amount of data to train. Transfer learning has been introduced to address this problem, but its effectiveness is sometimes marginal because the model does not effectively leverage prior knowledge. In this study, we propose a stochastic initial state randomization(SISR) method to enable robust knowledge transfer that promote generalized and sufficient knowledge transfer. We developed a simulation environment involving a cooperative robot transportation task. Experimental results show that successful tasks are achieved when SISR is applied, while tasks fail when SISR is not applied. We also analyzed how the amount of state information collected by the agents changes with the application of SISR.

• International Journal of Contents
• /
• v.4 no.4
• /
• pp.18-23
• /
• 2008

With the development of information communication and network technologies, ubiquitous era that supports various services regardless of places and time has been advancing. The development of such technologies has a great influence on educational environments. As a result, e-learning concepts that learners use learning contents in anywhere and anytime have been proposed. The various learning contents authoring systems that consider the e-learning environments have also been developed. However, since most of the existing authoring systems support only PC environments, they are not suitable for various ubiquitous mobile devices. In this paper, we design and implement a contents authoring system based on learning activities for u-learning environments. Our authoring system significantly improves the efficiency for authoring contents and supports various ubiquitous devices as well as PCs.

• Journal of the Korea Safety Management & Science
• /
• v.12 no.3
• /
• pp.271-281
• /
• 2010

In an industrial Era, OJT(On-the-Job Training) has been accepted as the field learning. But in a breaking up era, traditional field training needs to change and make an evolutionary model. Also, we need to make evolutionary model for various changing ways and means and need means to maximize the transformation of learning by operating learning organization. In knowledge based society, as people work and learn new knowledge in order to pass the experience knowledge and capabilities, they are not the traditional relationship between trainer and trainee but maximize work and learning, development and performance through several different ways. So, the study about new learning model is needed because the learning is creating the value and makes low cost and high efficiency about the elements of cost and time. We study the evolutionary model, OJF(On-the-Job Facilitating) - new learning methodology - through operating learning organization in S Electronics and its application practices.

• Proceedings of the KIEE Conference
• /
• 2003.07d
• /
• pp.2540-2542
• /
• 2003

This paper presents an internet-based self-learning educational system which can be enhancing efficiency in the learning process of Java language. The proposed self-learning educational system is called Java Web Player(JWP), which is a Java application program and is executable through Java Web Start technologies. In this paper, three important sequential learning processes : concept learning process, programming practice process and assessment process are integrated in the proposed JWP using Java Web Start technologies. This JWP enables the learners to achieve efficient and interesting self-learning since the learning process is designed to enhance the multimedia capabilities on the basis of educational technologies. Also, online voice presentation and its related texts together with moving images are synchronized for efficient language learning process. Furthermore, a simple/useful compiler is included in the JWP for providing language practice environment such as coding, editing, executing and debugging Java source files. Finally repeated practice can make the learners to understand easily the key concepts of Java language. 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.

• Journal of Multimedia Information System
• /
• v.9 no.1
• /
• pp.51-60
• /
• 2022

This paper raised the need to examine how the online education environment triggered by COVID-19 and the smart learning environment can be established in consideration of the improvement of education and learning through learning analysis. Many studies are being conducted in Korea, and the Ministry of Education is continuously striving to build a smart school by promoting strategies for promoting smart education on the way to a talent powerhouse. Nevertheless, there is no unified definition of smart learning, and it can be seen as customized (individualized) learning using smart devices. However, most of the discussions on the construction of smart schools so far have limitations in that they are limited to physical spaces. Accordingly, the opinions of teachers and learners were not sufficiently reflected in the establishment of the facility. This study intends to study smart learning in various departments. In addition, the subjects students in charge of the co-researcher of this study were analyzed. The total number of subjects was 951, and 434 responded to this study survey. In addition, students were well accepting the online environment, and in the future, regardless of COVID-19, research will be presented to improve mutual communication between professors and students in smart learning.

• Journal of Korean Society for Quality Management
• /
• v.42 no.3
• /
• pp.407-424
• /
• 2014

Purpose: The purpose of this study was to propose supply chain management efficiency by analyzing moderating effect on the training support project in small and midium-sized enterprise. Methods: The collected data through the survey were analyzed using multi?regression analysis. The measurement tools used for this study were divided into three dimensions such as personal characteristics, supportive environment and training effect. Results: The results of this study are as follows; first, it was found that the effects of personal characteristics and supportive environment were significant on training effect. Second, significant differences were found between supply chain management active group and inactive group. Third, moderating effect of supply chain manage efficiency was found between variables. Conclusion: SME supporting project in training program needs supply chain management efficiency for training effect. Supply chain management needs to be promoted in SMEs not only for performance but also for their workplace learning and learning culture.

• Journal of Convergence for Information Technology
• /
• v.7 no.1
• /
• pp.87-92
• /
• 2017

This paper discusses the possibility of enhancing the efficiency in Second Language Learning with the help of an educational information and content search system that utilizes a Cloud system based Big Data. The proposed system plays a role in tracing the exact information that user request considering the interest, level, and aptitude of each individual. Also, the system screen outs unnecessary learning materials if they do not correspond to the user level which in result is one of the key factor in the proposed system. Since it requires multiple steps of verification in providing the extracted results finding out a way to reduce the steps to shorten the time of process.

• Journal of the Korea Convergence Society
• /
• v.11 no.11
• /
• pp.103-114
• /
• 2020

This paper's research objective is to test educational content with different dimensions of knowledge during a graphic basic design course, while also proposing teaching plans and opinions for courses in flipped classrooms as well as enhancing educational efficiency. It categorizes educational content of courses based on the dimensions of knowledge in the learning objectives of Bloom's taxonomy. 120 students are divided into four experimental groups to respectively under go flipped classroom learning by using different dimensions of knowledge involved in course content. Course pretests and post tests are used to obtain and analyze experimental data. Among this knowledge, factual and conceptual knowledge obtained during extra curricular independent learning as well as programmed and meta-cognitive knowledge obtained during in-class learning from a flipped classroom can stimulate student's learning initiative and also enhance learning efficiency. According to research results and student feedback, this paper will propose targeted categorization methods for course content and also suggest educational strategies for these courses' flipped classroom model.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.25 no.8
• /
• pp.1032-1038
• /
• 2021

Rate of penetration(ROP) is one of the important variables for maximizing the drilling performance. In order to maximize drilling efficiency, it is necessary to increase the drilling speed, and real-time ROP prediction is important so that the driller can identify problems during drilling. The ROP has a high correlation with the drillstring rotational speed, weight on bit, and flow rate. In this paper, the ROP was predicted using a data-driven supervised learning model trained from the drilling efficiency parameters. As a result of comparison through the performance evaluation metrics of the regression model, the root mean square error(RMSE) of the RF model was 4.20 and the mean absolute percentage error(MAPE) was 9.08%, confirming the best predictive performance. The proposed method can be used as a base model for ROP prediction when constructing a real-time drilling operation guide system.