• KIPS Transactions on Computer and Communication Systems
• /
• v.10 no.5
• /
• pp.155-162
• /
• 2021

n-step TD learning is a combination of Monte Carlo method and one-step TD learning. If appropriate n is selected, n-step TD learning is known as an algorithm that performs better than Monte Carlo method and 1-step TD learning, but it is difficult to select the best values of n. In order to solve the difficulty of selecting the values of n in n-step TD learning, in this paper, using the characteristic that overestimation of Q can improve the performance of initial learning and that all n-step returns have similar values for Q ≈ Q^*, we propose a new learning target, which is composed of the maximum and the mean of all k-step returns for 1 ≤ k ≤ n. Finally, in OpenAI Gym's Atari game environment, we compare the proposed algorithm with n-step TD learning and proved that the proposed algorithm is superior to n-step TD learning algorithm.

• 한국정보교육학회:학술대회논문집
• /
• 2004.08a
• /
• pp.668-676
• /
• 2004

미술영역 중 서예영역은 7차 교육과정으로 개편되면서 $3{\sim}5$학년 각 8시간, 6학년 6시간으로 한 단원씩 배당되어 6차에 비해 시간수가 11% 감소되었다. 그로 인하여 각 단계별 학습내용을 체계적으로 학습할 시간이 부족하며, 지도할 자료 부족과 자료제작의 어려움, 또한 학생 개인 수준 차이로 서예지도에 많은 어려움이 있다. 이러한 문제를 해결하고자 본 논문에서는 초등 미술과 서예영역의 교육과정을 분석하여 각 학년별 교과 내용의 효과적인 필법 자료를 단계별로 학습의 흐름에 맞추어 투입할 수 있도록 다양한 멀티미디어 자료들을 본 학습 자료와 심화 학습 자료를 재구성하거나 개발하여 적용하는 초등학생을 위한 웹 기반 서예교육 시스템을 설계하고자 한다. 이를 통해 서예 지도를 위한 교사들의 자료 준비와 부족에 대한 부담감이 줄고, 학생 개인차를 고려하여 자료를 투입할 수 있어 자기 학습력 신장을 엿볼 수 있으며, 학생 스스로도 서예에 대한 흥미도가 높아져 향상된 작품을 볼 수 있을 것이다.

• Therapeutic Science for Rehabilitation
• /
• v.7 no.1
• /
• pp.51-62
• /
• 2018

Objective : This study was to verify the effects of occupation-based training with errorless learning and spaced retrieval on task learning and satisfaction of elderly with mild vascular dementia. Methods : The subjects of this study were 3 geriatric individuals with mild vascular dementia, ABA' + follow-up design was used. Intervention period was provided occupation-based training with errorless learning and spaced retrieval. The dependent variable was converted to ability of task performance every session and satisfaction before and after intervention. Result analysis was suggested through visual analysis and bar graph. Results : After the occupation-based training with errorless learning and spaced retrieval, Ability of task performance and satisfaction improved for all subjects. Conclusions : From this study, occupation-based training with errorless learning and spaced retrieval training was found to be an effective mediation for improving independence of task and satisfaction for people with mild dementia. This study could provide evidence for clinical application for occupation-based training.

• Journal of The Korean Association of Information Education
• /
• v.9 no.4
• /
• pp.595-606
• /
• 2005

7차교육과정에서의 초등 컴퓨터교육은 재량활동시간 또는 특별활동시간에 단순한 응용프로그램(워드프로세서, 엑셀, 파워포인트 등)을 활용하거나 인터넷을 이용한 정보검색 수준이다. 따라서, 문제해결 능력이나 논리적 사고력을 향상시키는데는 미흡하다. 인지능력이 뛰어난 초등학교 5, 6학년에는 이러한 컴퓨터 활용방법 교육보다 컴퓨터 기초원리 또는 프로그래밍교육이 필요하다. 본 논문은 프로그래밍교육을 재량 특별활동 시간에 학년별(5, 6학년) 또는 통합학년으로 학습할 수 있게 프로그래밍교육과정 60차시 분을 개발하여 제안하였다. 개발한 교육과정은 교수 및 학습자의 요구사항을 최대한 반영하기 위해 설문조사 등을 통해 얻은 기초자료를 참조하였다. 제안한 교육과정이 초등학교 5, 6학년에 적합하고 우수한지를 검토하기 위해 일부 차시의 교육과정으로 5, 6학년을 대상으로 재량활동시간에 직접 교수 학습한 후 자기평가, 설문조사를 실시하였다. 결과로 제안한 교육과정이 적합성과 우수성이 있다고 판단되었다.

• 한국정보교육학회:학술대회논문집
• /
• 2008.01a
• /
• pp.263-268
• /
• 2008

학교의 학습 과정에 있어서 학습자 간의 학습 능력의 차이는 존재하며 이를 해결하기 위해 교육과정에서는 개별 학습과 수준별 학습을 권장한다. 유무선 인터넷을 통한 수준별 학습은 최근에 많은 연구가 되고 있는 u-러닝(Ubiquitous Learning) 환경에도 부합하며 학습자 개개인이 자신의 속도와 수준에 맞게 자기주도적으로 학습을 하기에 알맞은 방법이라 할 수 있다. 이에 본 연구에서는 학습자가 시간과 공간에 구애받지 않고 자율적으로 수준에 맞게 학습할 수 있는 수준별 학습 시스템을 설계하였다. 특히 시스템에 체계성을 더하기 위해 개별화 학습 체제 중에서 과거 많은 연구를 통해 그 효과성이 입증된 Keller의 PSI(Personal System of Instruction) 이론을 활용하여 시스템의 각 과정을 설계하였다. 본 시스템의 장점은 다음과 같다. 첫째, 학습자가 원하는 시간과 공간에서 자신의 속도에 맞게 학습할 수 있으므로 자기주도적인 학습 능력을 기를 수 있다. 둘째, 시스템 구성상 평가를 통해 일정한 기준에 미달하면 목표에 도달할 때까지 계속 학습하고 도전해야 하므로 궁극적으로 완전학습에 도달할 수 있다. 셋째, 제한된 교실 상황에서 벗어나 온라인에서의 학습 지원이 가능하므로 학습자의 개인차에 따른 수준별 학습을 관리하고 책임져야 하는 교사의 부담을 덜어준다.

• Journal of Korean Home Economics Education Association
• /
• v.17 no.3 s.37
• /
• pp.123-138
• /
• 2005

The purposes of this study were to develop ICT time management program for middle school students using in the creative extra curricular activities and to test the effect of this program in the knowledge and attitude aspects of time management. The contents of a middle school Home Economics curriculum and nine different textbooks were analyzed Based on these analyses 5 ICT lesson plans on time management were developed. The ICT instructional method was used to develop 5 lesson plans. The second grade middle school students as an experimental group participated in 5 ICT time management classes during middle school creative extra curricular activities. The control group of students who had some condition with the experimental group were taught by traditional instructional methods. After experiment, the changes in attitude and knowledge of time management of both groups were analyzed using Analysis of Covaziance. The significant improvements of attitude and knowledge on time management were found among the experimental group of students compared with control group of students. Thus the following conclusion is made the ICT time management instruction conducted in this study was found loaming more effective than traditional one in attitudes and knowledge of time management

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.35S no.6
• /
• pp.81-87
• /
• 1998

This paper proposes an efficient face recognition system using phase difference between the face images. We use a Karhunen-Loeve transform for image compression and reconstruction, and obtain the phase difference by using normalized inner product of the two compressed images. The proposed system is rotation and light-invariant due to using the normalized phase difference, and somewhat shift-invariant due to applying the cosine function. The faster recognition than the conventional system and incremental training is possible in the proposed system. Simulations are conducted on the ORL images of 40 persons, in which each person has 10 facial images, and the result shows that the faster recognition than conventional recognizer using convolution network under the same recognition error rate of 8% does.

• Journal of Internet Computing and Services
• /
• v.21 no.5
• /
• pp.1-7
• /
• 2020

Deep neural networks(DNN), which are used as approximation functions in reinforcement learning (RN), theoretically can be attributed to realistic results. In empirical benchmark works, time difference learning (TD) shows better results than Monte-Carlo learning (MC). However, among some previous works show that MC is better than TD when the reward is very rare or delayed. Also, another recent research shows when the information observed by the agent from the environment is partial on complex control works, it indicates that the MC prediction is superior to the TD-based methods. Most of these environments can be regarded as 5-step Q-learning or 20-step Q-learning, where the experiment continues without long roll-outs for alleviating reduce performance degradation. In other words, for networks with a noise, a representative network that is regardless of the controlled roll-outs, it is better to learn MC, which is robust to noisy rewards than TD, or almost identical to MC. These studies provide a break with that TD is better than MC. These recent research results show that the way combining MC and TD is better than the theoretical one. Therefore, in this study, based on the results shown in previous studies, we attempt to exploit a random balance with a mixture of TD and MC in RL without any complicated formulas by rewards used in those studies do. Compared to the DQN using the MC and TD random mixture and the well-known DQN using only the TD-based learning, we demonstrate that a well-performed TD learning are also granted special favor of the mixture of TD and MC through an experiments in OpenAI Gym.

• Journal of the Korean Society of Child Welfare
• /
• no.56
• /
• pp.101-132
• /
• 2016

Fandom is a meaningful social phenomenon that represents a significant dimension of contemporary Korean youth culture. Previous research, however, has considered fandom as a deviant activity that is negative in nature, and there is scant research on the positive effects of fandom activity. This study examined the effects of off-line and on-line fandom activity participation on adolescents' academic time management capability. Data were collected from three waves (2010, 2012, and 2014) of the Korean Children and Youth Panel Survey (middle school cohort; N=2,206 at baseline). Results from the multi-level growth curve analysis were as follows. First, more than half of the adolescents reported participating in fandom activities, and the majority of them participated in both online and offline activities. Second, adolescents' academic time management capability improved over time. Third, both off-line and on-line fandom activity participation was positively associated with adolescents' academic time management capability at the baseline (first year of middle school). The size of this relationship, however, decreased over time (from middle school to high school). This study presented the importance of appropriate guidance by adults to develop and promote the positive aspects of fandom activity, and highlighted the need for further research that focuses on the positive aspects of fandom activities.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.665-666
• /
• 2008

본 연구는 학습 성취도 예측을 통한 완전학습 시스템을 연구하는데 그 목적이 있다. 학급 내의 95%의 학생들이 학습 과제의 90% 이상을 완전히 학습해 내는 것이 완전학습이다. 그러나 개인의 수준차로 인한 완전학습 도달 시간이 상이하고, 그 도달 시간을 파악하기가 어려우므로 현실적으로 완전학습에 도달하기란 쉬운 일이 아니다. 본 연구에서는 이러한 현실적인 어려움을 극복하고자 학생들의 과거 현재 학습 성취 데이터를 분석하여, 미래 학습 성취도를 예측함으로써 보충학습이 필요한 학생을 미리 선별하고, 학생별 특성과 수준에 맞는 보충학습 자료 제공을 통한 재학습 유도로 정해진 기간 내에 단계별(단원별, 학기별)로 완전학습에 도달할 수 있도록 하였다.