• 디지털융복합연구
• /
• 제6권3호
• /
• pp.13-22
• /
• 2008

The purpose of this study is to explore 'Flow Experience' of those studying English speaking and listening in the e-Learning environment. The exploration of flow experience in this study is based on the literature research of Csikszentmihalyi's flow models and other studies. There have been many studies on flow experience focusing on arts, leisure and sports in accordance with Csikszentmihalyi's original theory, however, his flow theory has recently been adapted to the educational field. Nonetheless, it is in the e-learning environment, rather than the face-to-face traditional teaming environment, that there is not enough flow state measurement scale. Therefore, it is important to develop as a stepping stone a flow state scale for those who study English speaking and listening by the cyber-native-speaker on e-Learning environment to improve their satisfaction and achievement.

• 한국지능시스템학회논문지
• /
• 제14권3호
• /
• pp.363-368
• /
• 2004

In this paper, the robustness property of 2nd-order iterative learning control(ILC) method for a class of linear and nonlinear discrete-time dynamic systems is studied. 2nd-order ILC method has the PD-type learning algorithm based on both time-domain performance and iteration-domain performance. It is proved that the 2nd-order ILC method has robustness in the presence of state disturbances, measurement noise and initial state error. In the absence of state disturbances, measurement noise and initialization error, the convergence of the 2nd-order ILC algorithm is guaranteed. A numerical example is given to show the robustness and convergence property according to the learning parameters.

• 한국정보처리학회논문지
• /
• 제6권3호
• /
• pp.633-640
• /
• 1999

Reinforcement learning performs learning through interacting with trial-and-error in dynamic environment. Therefore, in dynamic environment, reinforcement learning method like Q-learning and TD(Temporal Difference)-learning are faster in learning than the conventional stochastic learning method. However, because many of the proposed reinforcement learning algorithms are given the reinforcement value only when the learning agent has reached its goal state, most of the reinforcement algorithms converge to the optimal solution too slowly. In this paper, we present COMREL(COMpressed REinforcement Learning) algorithm for finding the shortest path fast in a maze environment, select the candidate states that can guide the shortest path in compressed maze environment, and learn only the candidate states to find the shortest path. After comparing COMREL algorithm with the already existing Q-learning and Priortized Sweeping algorithm, we could see that the learning time shortened very much.

• 한국컴퓨터그래픽스학회논문지
• /
• 제30권3호
• /
• pp.109-123
• /
• 2024

딥러닝과 강화학습을 활용한 비전 기반 엔드투엔드 자율주행 시스템 관련 연구가 지속적으로 증가하고 있다. 일반적으로 이러한 시스템은 위치, 속도, 방향, 센서 데이터 등 연속적이고 고차원적인 차량의 상태를 잠재 특징 벡터로 인코딩하고, 이를 차량의 주행 정책으로 디코딩하는 두 단계로 구성된다. 도심 주행과 같이 다양하고 복잡한 환경에서는 Variational Autoencoder(VAE)나 Convolutional Neural Network(CNN)과 같은 네트워크를 이용한 효율적인 상태 표현 방법의 필요성이 더욱 부각된다. 본 논문은 차량의 이미지 상태 표현이 강화학습 성능에 미치는 영향을 분석하였다. CARLA 시뮬레이터 환경에서 실험을 수행하였고, 차량의 전방 카메라 센서로부터 취득한 RGB 이미지 및 Semantic Segmented 이미지를 각각 VAE와 Vision Transformer(ViT) 네트워크로 특징 추출하여 상태 표현 학습에 활용하였다. 이러한 방법론이 강화학습에 미치는 영향을 실험하여, 데이터 유형과 상태 표현 기법이 자율주행의 학습 효율성과 결정 능력 향상에 어떤 역할을 하는지를 실험하였다.

• International Journal of Computer Science & Network Security
• /
• 제22권9호
• /
• pp.300-306
• /
• 2022

The theoretical analysis and synthesis of prospects for the development of distance learning in Ukraine, the main topical problems of distance education in Ukraine are considered, the main factors that hinder the introduction of distance learning are analyzed, to pay attention to the need to increase the level of computer literacy among Ukrainian educators and the formation of modern methodology of distance learning, in particular, a single, systematic, national approach of organization, coordination and control in this area. Research methods: analytical method, method of structural and functional analysis, phenomenological method, content analysis method, philosophical reflection method, sociological methods (questionnaire, interview).

• 과학교육연구지
• /
• 제35권2호
• /
• pp.230-239
• /
• 2011

이 연구는 중학교 일기변화 단원의 개념 검사지를 개발하여 지식상태 분석법으로 일기변화 관련 개념의 위계를 분석하였고, 개별 학습자의 지식상태와 위계에 따른 수업효과를 알아보았다. 일기변화 관련 개념에 대하여 중학생들이 갖고 있는 지식상태의 위계구조는 '습도 ${\rightarrow}$ 기단 ${\rightarrow}$ 구름 ${\rightarrow}$ 강수 ${\rightarrow}$ 전선 ${\rightarrow}$ 일기' 순이었다. 일기변화 개념 검사에서 같은 점수를 획득하여 학습능력이 비슷할 것으로 추정되는 개별 학습자의 지식상태가 서로 다르게 나타나는 사례가 있었다. 즉, 지식상태의 구조화 정도가 달랐는데, 이는 학습자의 지식상태 구조에 따라 서로 다른 교수-학습 처방이 이루어져야 함을 시사한다. 학습자의 지식상태 분석은 개별 학습자마다의 학습 처방과 선수학습평가의 역할을 할 수 있다. 학습자의 지식상태에 따른 개념의 위계를 고려한 교수-학습 효과를 알아보기 위해 교과서의 내용 제시 순서와 비교하여 수업한 결과, 학습자의 지식상태를 고려하여 수업한 경우에 개념의 성취도가 유의미(p < .05)하게 더 높았다. 이는 교사들이 일기변화 단원을 지도할 때, 학습자의 지식상태를 파악하여 교육과정을 토대로 교과서의 학습 내용을 재순서화 함으로써 더 효과적인 교수-학습이 이루어질 수 있음을 보여 준다.

• Journal of Electrical Engineering and Technology
• /
• 제11권2호
• /
• pp.425-436
• /
• 2016

In this paper, a parameter optimization based iterative learning control strategy is presented for permanent magnet synchronous motor control. This paper analyzes the mechanism of iterative learning control suppressing PMSM torque ripple and discusses the impact of controller parameters on steady-state and dynamic performance of the system. Based on the analysis, an optimization problem is constructed, and the expression of the optimal controller parameter is obtained to adjust the controller parameter online. Experimental research is carried out on a 5.2kW PMSM. The results show that the parameter optimization based iterative learning control proposed in this paper achieves lower torque ripple during steady-state operation and short regulating time of dynamic response, thus satisfying the demands for both steady state and dynamic performance of the speed regulating system.

• International Journal of Computer Science & Network Security
• /
• 제22권6호
• /
• pp.251-255
• /
• 2022

The article considers the state and trends of distance learning in the world and Ukraine, identifies the main species differences between distance education and other forms of education, analyzes the state of the global market for educational services provided via the Internet. Important features and characteristics of distance learning, examples of its organization in higher education, as well as statistics on the development of distance learning in our country. The main problematic points on the way to the implementation of the distance education system in Ukraine and the factors that hinder the development of this promising form of education are outlined.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1990년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 26-27 Oct. 1990
• /
• pp.963-968
• /
• 1990

An iterative learning control scheme for tracking control of a class of uncertain nonlinear systems is presented. By introducing a model reference adaptive controller in the learning control structure, it is possible to achieve zero tracking of unknown system even when the upperbound of uncertainty in system dynamics is not known apriori. The adaptive controller pull the state of the system to the state of reference model via control gain adaptation at each iteration, while the learning controller attracts the model state to the desired one by synthesizing a suitable control input along with iteration numbers. In the controller role transition from the adaptive to the learning controller takes place in gradually as learning proceeds. Another feature of this control scheme is that robustness to bounded input disturbances is guaranteed by the linear controller in the feedback loop of the learning control scheme. In addition, since the proposed controller does not require any knowledge of the dynamic parameters of the system, it is flexible under uncertain environments. With these facts, computational easiness makes the learning scheme more feasible. Computer simulation results for the dynamic control of a two-axis robot manipulator shows a good performance of the scheme in relatively high speed operation of trajectory tracking.

• Asia pacific journal of information systems
• /
• 제32권2호
• /
• pp.226-248
• /
• 2022

The existing model for sentiment analysis of product reviews learned from past data and new data was labeled based on training. But new data was never used by the existing system for making a decision. The proposed Aspect-based multi-agent Deep Reinforcement learning Sentiment Analysis (ADRSA) model learned from its very first data without the help of any training dataset and labeled a sentence with aspect category and sentiment polarity. It keeps on learning from the new data and updates its knowledge for improving its intelligence. The decision of the proposed system changed over time based on the new data. So, the accuracy of the sentiment analysis using deep reinforcement learning was improved over supervised learning and unsupervised learning methods. Hence, the sentiments of premium customers on a particular site can be explored to other customers effectively. A dynamic environment with a strong knowledge base can help the system to remember the sentences and usage State Action Reward State Action (SARSA) algorithm with Bidirectional Encoder Representations from Transformers (BERT) model improved the performance of the proposed system in terms of accuracy when compared to the state of art methods.