• International Journal of Internet, Broadcasting and Communication
• /
• 제16권2호
• /
• pp.161-168
• /
• 2024

As the size of big data models grows, distributed training is emerging as an essential element for large-scale machine learning tasks. In this paper, we propose ParamHub for distributed data training. During the training process, this agent utilizes the provided data to adjust various conditions of the model's parameters, such as the model structure, learning algorithm, hyperparameters, and bias, aiming to minimize the error between the model's predictions and the actual values. Furthermore, it operates autonomously, collecting and updating data in a distributed environment, thereby reducing the burden of load balancing that occurs in a centralized system. And Through communication between agents, resource management and learning processes can be coordinated, enabling efficient management of distributed data and resources. This approach enhances the scalability and stability of distributed machine learning systems while providing flexibility to be applied in various learning environments.

• 한국문헌정보학회지
• /
• 제38권4호
• /
• pp.85-104
• /
• 2004

본 연구는 인터넷 기반의 교육정보서비스 기관에서 채택하고 있는 학습객체의 개념을 도입하여 학교도서관 정보시스템(DLS)의 메타데이터에 교육 관련 요소를 추가, 확장할 수 있는 방안을 제시하고자 한다. 교수${\cdot}$학습의 상황에 따라서 정보자원에 접근토록 하고, 정보자원에 포함된 내부 데이터 요소의 제반 특성을 메타데이터로 기술하고 색인 함으로써 학습객체 단위로 접근할 수 있는 방안을 제시하는데 목적이 있다. 기존의 DLS의 메타데이터에 , , 요소를 추가하여 교수${\cdot}$학습 상황에 따라서 정보자원에 접근할 수 있도록 하고 요소를 이용하여 내부 학습객체의 개념을 수용할 수 있는 방안을 제시하였다.

• 전산구조공학
• /
• 제35권4호
• /
• pp.27-33
• /
• 2022

• Journal of Information Technology Applications and Management
• /
• 제17권3호
• /
• pp.135-150
• /
• 2010

Interactivity has been referred to as an important element promoting students' active participation in virtual classes. Assuming that interactivity cannot be defined by a single dimension, this study proposes multi-dimensional interactivity. Multi-dimensional interactivity includes all types of interactivity in e-learning. This study explored multi-dimensional interactivity which affects learners' satisfaction with e-learning. Data were collected from 132 students who had attended e-learning courses and the relationship between multi-dimensional interactivity and learners' satisfaction levels were tested through regression analysis. The result of this study showed that mechanical, reactive, and creative interactivity were positively related to learners' satisfaction. However, social interactivity seemed not to be related to learners' satisfaction. This study provides new insights on interactivity and verifies the importance of the multi-dimensional interactivity. The result of this study is expected to provide practical implications for interactivity strategies in e-learning.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• 제6권4호
• /
• pp.271-276
• /
• 2006

To implement a generalization of value functions in Adaptive Search Element (ASE)-reinforcement learning, CMAC (Cerebellar Model Articulation Controller) is integrated into ASE controller. ASE-reinforcement learning scheme is briefly studied to discuss how CMAC is integrated into ASE controller. Neighbourhood Sequential Training for CMAC is utilized to establish the look-up table and to produce discrete control outputs. In computer simulation, an ASE controller and a couple of ASE-CMAC neural network are trained to balance the inverted pendulum on a cart. The number of trials until the controllers are established and the learning performance of the controllers are evaluated to find that generalization ability of the CMAC improves the speed of the ASE-reinforcement learning enough to realize the cartpole control system.

• Structural Engineering and Mechanics
• /
• 제77권1호
• /
• pp.47-56
• /
• 2021

This paper deals with damage detection using a Gapped Smoothing Method (GSM) combined with deep learning. Convolutional Neural Network (CNN) is a model of deep learning. CNN has an input layer, an output layer, and a number of hidden layers that consist of convolutional layers. The input layer is a tensor with shape (number of images) × (image width) × (image height) × (image depth). An activation function is applied each time to this tensor passing through a hidden layer and the last layer is the fully connected layer. After the fully connected layer, the output layer, which is the final layer, is predicted by CNN. In this paper, a complete machine learning system is introduced. The training data was taken from a Finite Element (FE) model. The input images are the contour plots of curvature gapped smooth damage index. A free-free beam is used as a case study. In the first step, the FE model of the beam was used to generate data. The collected data were then divided into two parts, i.e. 70% for training and 30% for validation. In the second step, the proposed CNN was trained using training data and then validated using available data. Furthermore, a vibration experiment on steel damaged beam in free-free support condition was carried out in the laboratory to test the method. A total number of 15 accelerometers were set up to measure the mode shapes and calculate the curvature gapped smooth of the damaged beam. Two scenarios were introduced with different severities of the damage. The results showed that the trained CNN was successful in detecting the location as well as the severity of the damage in the experimental damaged beam.

• Composites Research
• /
• 제36권3호
• /
• pp.205-210
• /
• 2023

본 연구에서는 폼 구조의 효율적인 유효 기계적 물성 및 열전도율 예측을 위한 균질화 데이터 기반 전이학습 프레임워크를 개발하였다. Eshelby 텐서 기반의 평균장 균질화(Mean-field homogenization, MFH)는 타원체 형태의 공동을 포함하는 다공성 구조의 물성을 효율적으로 예측할 수 있지만, 셀룰러(cellular) 폼 구조의 물성은 정확하게 예측하기 어렵다. 한편, 유한요소 균질화(Finite element homogenization, FEH)는 정확성은 높지만 상대적으로 높은 해석 시간을 동반한다. 본 논문에서는 평균장 균질화와 유한요소 균질화의 장점을 결합한 데이터 기반 전이학습 프레임워크(Framework)를 제안하였다. 구체적으로, 대량의 평균장 균질화 데이터를 도출하여 사전학습 모델(Pre-trained model)을 구축하고, 상대적으로 소량의 유한요소 균질화 데이터를 이용하여 미세 조정(Fine-tuning) 하였다. 제안된 프레임워크를 검증하기 위한 수치 예제를 수행하였으며, 해석 정확도를 확인하였다. 본 연구의 결과는 다양한 폼 구조를 가진 재료의 해석에 적용할 수 있을 것으로 기대한다.

• 한국지능시스템학회논문지
• /
• 제10권5호
• /
• pp.466-477
• /
• 2000

이 논문에서는 운반차-막대 시스템을 제어하기 위한 CMAC을 이용한 적응 학습 제어계를 개발하기 위하여, 적응비평학습을 이용하는 신경망 제어계에 관한 여러 연구 문헌들을 조사하고, ASE 요소를 이용하는 적응비평학습 기법을 CMAC을 바탕으로 하는 제어계에 통합하였다. 적응비평학습 기법을 CMAC에 구현하는데 있어서의 변환 문제를 검토하고, CMAC 제어계와 ASE 제어계가 운반차-막대 문제를 학습하는 속도를 비교하여, CMAC 제어계의 학습 속도가 빠르기는 하지만, 입력 공간의 더 넓은 영역에 대해서는 학습효과를 발휘하지 못하는 문제의 관점에서 적응비평학습 방법이 CMAC의 특성과 어울리는지를 고찰하였다.

• 대한간호
• /
• 제31권5호
• /
• pp.65-71
• /
• 1992

In nursing, a practice discipline, it has been argued whether the mastery of clinical competence depends on types of learning styles, learning experience, and the use of specialized skills. All these problems are not limited to nursing education. Later educators identified the importance of reflective learning which is a vital element in any form of learning and that teachers and students need to consider how they can incorporate some forms of reflection in the courses. The purpose of this study is to review educational articles for understanding better what reflection in learning is, to identify the theme which is of important relevance to professional practice, from the book, Educating the reflective practitioner, and to discuss the theme within nursing education. Reflection in learning was defined by Dewey(1933) as the process which is involved the perception of relationships and connections between the parts of an experience. This experience is passed on when two people becoming involved with each other in a conversation. schon(1987) emphasized that learning conversation, which is a part of the interaction of student and teacher, is an important factor of the process of reflection-om-actopm. In clinical nursing education, good relationships between teacher and student, faculty's role, interpersonal skills are critical in learning conversation. Then Practing nurses who accept the need to choose nursing actionss on the basis of reflection, who accepet the necessity for understanding and being able to communicate the reasons for action are a powerfful force for the development of nursing into an increasingly more effective profession for the benefit of patients.

• Educational Technology International
• /
• 제10권2호
• /
• pp.127-143
• /
• 2009

The purpose of this study was to identify the interaction patterns in distance only mode e-Learning. In order to investigate this study, messages shown in the electronic notice board were analyzed to see how interaction occurs between teacher and learner or learner and learner under the e-learning of cyber university. To analyze messages was applied according to the framework by Henri's contents analysis model. As a result of contents analysis on electronic board, the participative dimension was 399 messages. A learner put on 7~8 messages a day. The number of messages was low compared to the number of learners, but the number of inquiries was about 140. That means that each learner contacts and checks messages at least once a day. The meaning dimension was 600 units. The main interaction patterns were Interactive-social-cognitive-metacognitive. This means that e-Learning in distance only mode leads a positive attitude of learners as a self-directed learning, and needs teacher's well-structured instructional strategies for increasing interaction. In conclusion, social dimension and interactive dimension of messages support learners psychologically in the process of learning though they directly guide learning under the circumstances of e-learning lacking face-to-face element. It can be interpreted that the teacher's role is significantly important in order to attract learners' positive participation and cognitive and meta-cognitive dimension of messages and activities