• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.164.1-164
• /
• 2001

Q-learning is a recent reinforcement learning algorithm that does not need a modeling of environment and it is a suitable approach to learn behaviors for autonomous agents. But when it is applied to multi-agent learning with many I/O states, it is usually too complex and slow. To overcome this problem in the multi-agent learning system, we propose the successive Q-learning algorithm. Successive Q-learning algorithm divides state-action pairs, which agents can have, into several Q-functions, so it can reduce complexity and calculation amounts. This algorithm is suitable for multi-agent learning in a dynamically changing environment. The proposed successive Q-learning algorithm is applied to the prey-predator problem with the one-prey and two-predators, and its effectiveness is verified from the efficient avoidance ability of the prey agent.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제43권3호
• /
• pp.291-307
• /
• 2004

This study focused on the developing a system of computer adaptive diagnosis and instruction(CADI). This system is a conceptual model that connected learning with assesment by using new media such as computers, multimedia, and new technologies. In this conceptual model, adaptive diagnosis means tailored or customized diagnostic evaluation, and adaptive instruction implies tailored or customized instruction. The connection between learning and assesment suggests that they are closely related to determine following learning contents and learning methods. CADI's expected effect are 1) it can contribute to real learning of core concept, 2) it can enlarge the educational opportunities, 3) it can help students study by student himself and learn media literacy, 4) information for evaluation functions more essential roles, 5) it is possible to work cooperatively with any other school subject.

• 전자공학회논문지B
• /
• 제29B권1호
• /
• pp.94-101
• /
• 1992

The multilayer perceptron, trained by the error back-propagation learning rule, has been known as a mapping network which can represent arbitrary functions. However depending on the complexity of a function and the initial weights of the multilayer perceptron, the error back-propagation learning may fall into a local minimum or a flat area which may require a long learning time or lead to unsuccessful learning. To solve such difficulties in training the multilayer perceptron by standard error back-propagation learning rule, the paper proposes a learning method which progressively enlarges the learning domain from a small area to the entire region. The proposed method is devised from the investigation on the roles of hidden nodes and connection weights in the multilayer perceptron which approximates a function of one variable. The validity of the proposed method was illustrated through simulations for a function of one variable and a function of two variable with many extremal points.

• East Asian mathematical journal
• /
• 제35권2호
• /
• pp.199-216
• /
• 2019

The purpose of this study is to analyze the needs of mathematics learning program using web or application for the development of adaptive mathematics learning program. For this study, a questionnaire survey was conducted for middle school students. The results of the 290 responses were analyzed to identify students' needs for adaptive mathematics learning. As a result, learners wanted functions to design their own learning and to support self-directed learning. The results of this study are expected to be used as basic data for the development of a adaptive mathematics learning program reflecting students' needs.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제49권3호
• /
• pp.353-373
• /
• 2010

This research was to help slow learners to be motivated and to make their outcome productive, using GSP based on the mathematization theory for learning mathematics, as a way of encouraging the learner-centered approach. With 2 of the second graders in a high school, who had not yet understood trigonometric functions in their first grade period, 7 units of lesson plans were designed for the research. The results showed that first, understanding real life contexts and analyzing properties by observation, and experiment using GSP, to build the concept of trigonometric functions could be a foothold on which learner's organization and outcome from a horizontal mathematization led to vertical mathematization. Despite the delay during the level-up-stage for a while, the learners could attain the vertical mathematization stage and moreover the applicative mathematization through effective use of GSP and the interaction between the learners or a teacher and the learners. Second, using GSP was a vertical tool of connecting horizontal mathematization with vertical mathematization in forming the concept of trigonometric functions and its meaning could be understood by their verbalizing and presenting the outcomes through their active performance. Using GSP is helpful for slow learners to overcome learning difficulties, based on the instructional materials designed by Realistic Mathematics Education.

• 한국수학교육학회지시리즈A:수학교육
• /
• 제58권1호
• /
• pp.79-99
• /
• 2019

This study is a study to collect information about 'Limitations of functions' related learning. Especially, this study was conducted on three students who can find answers by algebraic procedure in the process of extreme problem solving. Students have had the experience of converting from their algebraic procedures to graphical expressions. This shows how they reflect on their algebraic procedures. This study is a study that observes these parts. To accomplish this, twelfth were teaching experiment in three high school students. And we analyzed the contents related to the research topic of this study. Through this, students showed the difference of expressions in the method of finding limits by using algebraic interpretation methods and graphs. In addition, we examined the connectivity of the limitations of functions problem solving process of functions using algebraic procedures and graphs in the process of converting algebraic expressions to graph expressions. This study is a study of how students construct limit concepts. As in this study, it is meaningful to accumulate practical information about students' limit conceptual composition. We hope that this study will help students to study limit concept development process for students who have no limit learning experience in the future.

• 대한수학교육학회지:학교수학
• /
• 제19권3호
• /
• pp.593-615
• /
• 2017

본 연구에서는 학교 수학의 내용을 담아낸 모바일 기반 수학 학습 어플리케이션을 개발하고, 이를 수학 교수 학습 상황에서 활용할 수 있는 방안을 모색하였다. 먼저, 중학교 수학의 '이차함수와 그래프', '대푯값과 산포도' 단원에서 조작을 통한 개념 탐구와 토론 학습이 가능하도록 어플리케이션 <중3수학교실>을 안드로이드 버전으로 수학교육 전문가, 수학 교사, 컴퓨터공학 전문가, 디지털 애니메이션 감독과 협업하여 개발하였다. 이때, 예비교사의 도움을 받아 중학교 3학년 남녀 학생 4명을 대상으로 개발한 어플리케이션을 적용하여 그 활용가능성을 검증하였다. 또한, 수학 수업에 어플리케이션을 활용하는 교수 학습 지도안과 어플리케이션 활용 매뉴얼을 개발하고 중학교 3학년 1개 학급을 대상으로 사례 연구를 실시하여 그 현장 적용 가능성을 모색하였다.

• 정보처리학회논문지B
• /
• 제11B권6호
• /
• pp.717-722
• /
• 2004

본 논문은 가중 퍼지소속함수 기반 신경망(Neural Network with Weighted Fuzzy Membership Functions, NNWFM)을 이용하여 Wisconsin breast cancer의 예측을 수행하는 퍼지규칙을 추출하고 있다. NNWFM는 자기적응적(self adaptive)가중 퍼지소속함수를 가지고 주어진 입력 데이터로부터 학습하여 퍼지규칙을 생성하고 이론 기반으로 예측을 수행한다. 신경망 구조의 중간 부분인 하이퍼박스(hyperbox)들은 n개의 대, 중, 소의 가중 퍼지소속함수 집합으로 구성되며, 학습 후 각 집합은 퍼지집합의 bounded sum을 사용하여 다시 하나의 가중 퍼지소속함수로 합성된다. n개의 특징입력(feature input)은 학습된 모든 하이퍼박스에 연결되어 예측 작업을 수행한다. NNWFM으로 추출된 2개의 퍼지규칙은 99.41%의 예측 인식율을 가지며 이는 퍼지규칙의 수와 인식율에 있어 현재 발표된 논문의 결과보다 우수함을 보여준다.

• Asia pacific journal of information systems
• /
• 제5권2호
• /
• pp.143-176
• /
• 1995

We present an overview of the Extensional Object Model (ExOM) and describe in detail the learning and classification components which integrate concepts from machine learning and object-oriented databases. The ExOM emphasizes flexibility in information acquisition, learning, and classification which are useful to support tasks such as diagnosis, planning, design, and database mining. As a vehicle to integrate machine learning and databases, the ExOM supports a broad range of learning and classification methods and integrates the learning and classification components with traditional database functions. To ensure the integrity of ExOM databases, a subsumption testing rule is developed that encompasses categories defined by type expressions as well as concept definitions generated by machine learning algorithms. A prototype of the learning and classification components of the ExOM is implemented in Smalltalk/V Windows.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1999년도 제14차 학술회의논문집
• /
• pp.292-297
• /
• 1999

Evolving in artificial agent is an extremely difficult problem, but on the other hand, a challenging task. At present the studies mainly centered on single agent learning problem. In our case, we use simulated soccer to investigate multi-agent cooperative learning. Consider the fundamental differences in learning mechanism, existing reinforcement learning algorithms can be roughly classified into two types-that based on evaluation functions and that of searching policy space directly. Genetic Programming developed from Genetic Algorithms is one of the most well known approaches belonging to the latter. In this paper, we give detailed algorithm description as well as data construction that are necessary for learning single agent strategies at first. In following step moreover, we will extend developed methods into multiple robot domains. game. We investigate and contrast two different methods-simple team learning and sub-group loaming and conclude the paper with some experimental results.