• Nuclear Engineering and Technology
• /
• v.54 no.9
• /
• pp.3283-3292
• /
• 2022

Based on the Deep Q-Network(DQN) algorithm of reinforcement learning, an active fault-tolerance method with incremental action is proposed for the control system with sensor faults of the once-through steam generator(OTSG). In this paper, we first establish the OTSG model as the interaction environment for the agent of reinforcement learning. The reinforcement learning agent chooses an action according to the system state obtained by the pressure sensor, the incremental action can gradually approach the optimal strategy for the current fault, and then the agent updates the network by different rewards obtained in the interaction process. In this way, we can transform the active fault tolerant control process of the OTSG to the reinforcement learning agent's decision-making process. The comparison experiments compared with the traditional reinforcement learning algorithm(RL) with fixed strategies show that the active fault-tolerant controller designed in this paper can accurately and rapidly control under sensor faults so that the pressure of the OTSG can be stabilized near the set-point value, and the OTSG can run normally and stably.

• The Mathematical Education
• /
• v.53 no.1
• /
• pp.75-92
• /
• 2014

In this paper, we summarize briefly some of the most salient features of Repkina & Zaika's theory of learning action formation levels. We concretize Repkina & Zaika's theory by comparing various points of view of Uoo, Polya, Krutetskii, and Davydov et al. In this study we are able to diagnose students' learning action formation levels in the process of mathematics problem solving. In addition we use interview method to collect various information about students' levels. As a result we suggest data related with each level of learning action formation, and characteristics of students who belong to each level of learning action formation.

• Human Ecology Research
• /
• v.58 no.4
• /
• pp.561-583
• /
• 2020

This study developed and implemented a teaching · learning process plan for Home Economics in middle school by converging Habermas's three systems of action. It also examined the effect of the class through the evaluation of students and teachers who participated in the class. This study developed 10 sessions for a teaching and learning process plan by converging three systems of action and reconstructing learning elements related to 'balanced meal plan' and 'food choice' according to the practical action teaching model. After class, we surveyed the degree of help for students, analyzed the learning activity sheets, and analyzed the reflection journals of teachers to evaluate the effects of the class. This class was found to be the most helpful in practicing the healthy dietary life of students, expanding their thoughts, understanding learning contents, and helping them change their lives. As a result of analyzing the learning activity sheet, students gained enlightenment by reflecting and evaluating their action through the class; in addition, changes in interest, awareness, and action appeared. Through the convergence of three systems of action, teachers who practiced the class criticized and realized the act that students were unconsciously accepted. In addition, it confirmed the possibility that students could change their lives, family and society by promoting optimal nutrition and health for a good life that pursues the best good.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.8 no.3
• /
• pp.220-224
• /
• 2008

This paper presents the polygon-based Q-leaning and Parallel SVM algorithm for object search with multiple robots. We organized an experimental environment with one hundred mobile robots, two hundred obstacles, and ten objects. Then we sent the robots to a hallway, where some obstacles were lying about, to search for a hidden object. In experiment, we used four different control methods: a random search, a fusion model with Distance-based action making (DBAM) and Area-based action making (ABAM) process to determine the next action of the robots, and hexagon-based Q-learning, and dodecagon-based Q-learning and parallel SVM algorithm to enhance the fusion model with Distance-based action making (DBAM) and Area-based action making (ABAM) process. In this paper, the result show that dodecagon-based Q-learning and parallel SVM algorithm is better than the other algorithm to tracking for object.

• Asia-Pacific Journal of Business Venturing and Entrepreneurship
• /
• v.10 no.6
• /
• pp.99-106
• /
• 2015

This is a case study on organizational learning with action learning. It targets B industry belonging to Shoe manufacturer. We apply action learning techniques as consulting skills to promote the organization of specific learning activities. Action Learning solves the challenges faced by the company with the ability to enhance the member while participating in the program. Therefore, it is a good methodology to overcome the uncertainty environment. Through a case study, in the maturing process of a learning organization can see the conditions that are necessary for the ongoing maintenance of that identity, organizational learning activities. Findings to the continued operation of the enterprise learning organization suggest the establishment of a learning organization, and direction and strategic importance. Systems and learning environments should be built and then repeat the process of practice to master the new learning organization. It suggests to learn a new organizations operating methods that require repetition of the course of action.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.16 no.4
• /
• pp.270-275
• /
• 2016

We present an approach for acceleration actor-critic algorithm for reinforcement learning with continuous action space. Actor-critic algorithm has already proved its robustness to the infinitely large action spaces in various high dimensional environments. Despite that success, the main problem of the actor-critic algorithm remains the same-speed of convergence to the optimal policy. In high dimensional state and action space, a searching for the correct action in each state takes enormously long time. Therefore, in this paper we suggest a search accelerating function that allows to leverage speed of algorithm convergence and reach optimal policy faster. In our method, we assume that actions may have their own distribution of preference, that independent on the state. Since in the beginning of learning agent act randomly in the environment, it would be more efficient if actions were taken according to the some heuristic function. We demonstrate that heuristically-accelerated actor-critic algorithm learns optimal policy faster, using Educational Process Mining dataset with records of students' course learning process and their grades.

• The Journal of the Korea Contents Association
• /
• v.18 no.5
• /
• pp.525-538
• /
• 2018

Action learning is a teaching method that is taught through the process of finding real problems and finding solutions. The purpose of this study was to examine the effectiveness of action learning for the college course 'media education for children' which was offered by the Department of Child Psychology and Education at S university in Seoul. According to the process of action learning suggested by Marquardt(2000), college students developed the problem of children's news literacy education and set the purpose, goal and contents of news literacy and developed teaching and learning method for the 6th grade elementary students and applied it to the education field. The college students' evaluation about the action learning lesson were collected by questionnaire and verbal. They found that active and voluntary problem-solving and field application through action learning helped to achieve the goals of the children's media education class. Finally, the importance of close cooperation with the educational field and of appropriate feedback were discussed as a necessary condition of action learning teaching method.

• Journal of Korea Multimedia Society
• /
• v.13 no.12
• /
• pp.1820-1831
• /
• 2010

When interacting with unknown environments, an autonomous agent needs to decide which action or action order can result in a good state and determine the transition probability based on the current state and the action taken. The traditional multiple sequential learning model requires predefined probability of the states' transition. This paper proposes a multiple sequential learning and prediction system with definition of autonomous states to enhance the automatic performance of existing AI algorithms. In sequence learning process, the sensed states are classified into several group by a set of proposed motivation filters to reduce the learning computation. In prediction process, the learning agent makes a decision based on the estimation of each state's cost to get a high payoff from the given environment. The proposed learning and prediction algorithms heightens the automatic planning of the autonomous agent for interacting with the dynamic unknown environment. This model was tested in a virtual library.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.18 no.3
• /
• pp.291-296
• /
• 2008

This paper presents the dodecagon-based Q-leaning and SVM algorithm for object search with multiple robots. We organized an experimental environment with several mobile robots, obstacles, and an object. Then we sent the robots to a hallway, where some obstacles were tying about, to search for a hidden object. In experiment, we used four different control methods: a random search, a fusion model with Distance-based action making(DBAM) and Area-based action making(ABAM) process to determine the next action of the robots, and hexagon-based Q-learning and dodecagon-based Q-learning and SVM to enhance the fusion model with Distance-based action making(DBAM) and Area-based action making(ABAM) process.

• The Journal of the Korea Contents Association
• /
• v.16 no.12
• /
• pp.35-42
• /
• 2016

The purpose of this study is to examine the effect of action learning approaches on problem-solving skills and learning agency of nursing undergraduate students. This experimental study is designed for a nonequivalent control group. The program was put into practice 2 times a week for 4 weeks. The number of subjects in this research consists of 105, where 53 of the experimental group participated in action learning program and 52 of the control group didn't do. The data was analyzed by ${\chi}^2$-test, Chi-Square test, t-test and paired t-test. The effects of action learning approaches on learning outcomes in nursing process courses are as follows: The problem solving ability of the experimental group has been more elevated than that of the control group. The experimental group has made increase in self directed learning skills. The action learning approaches on learning outcomes in nursing process courses are convenient in nursing process courses. This study has significance in that it identified the availability of the action learning program and that it would be useful teaching and learning method to achieve learning outcomes.