• International Journal of Internet, Broadcasting and Communication
• /
• v.15 no.3
• /
• pp.166-175
• /
• 2023

Reinforcement learning can be applied to a wide variety of problems. However, the fundamental limitation of reinforcement learning is that it is difficult to derive an answer within a given time because the problems in the real world are too complex. Then, with the development of neural network technology, research on deep reinforcement learning that combines deep learning with reinforcement learning is receiving lots of attention. In this paper, two types of neural networks are combined with reinforcement learning and their characteristics were compared and analyzed with existing value-based reinforcement learning algorithms. Two types of neural networks are FNN and CNN, and existing reinforcement learning algorithms are SARSA and Q-learning.

• Journal of Institute of Control, Robotics and Systems
• /
• v.3 no.4
• /
• pp.350-356
• /
• 1997

It is desirable for autonomous robot systems to possess the ability to behave in a smooth and continuous fashion when interacting with an unknown environment. Although Q-learning requires a lot of memory and time to optimize a series of actions in a continuous state space, it may not be easy to apply the method to such a real environment. In this paper, for continuous state space applications, to solve problem and a triangular type Q-value model\ulcorner This sounds very ackward. What is it you want to solve about the Q-value model. Our learning method can estimate a current Q-value by its relationship with the neighboring states and has the ability to learn its actions similar to that of Q-learning. Thus, our method can enable robots to move smoothly in a real environment. To show the validity of our method, navigation comparison with Q-learning are given and visual tracking simulation results involving an 2-DOF SCARA robot are also presented.

• The Journal of the Korea Contents Association
• /
• v.20 no.12
• /
• pp.49-57
• /
• 2020

This study predicts the ratio of added value, which represents the competitiveness of export industries in South Korea, using various machine learning techniques. To enhance the accuracy and stability of prediction, forecast combination technique was applied to predicted values of machine learning techniques. In particular, this study improved the efficiency of the prediction process by selecting key variables out of many variables using recursive feature elimination method and applying them to machine learning techniques. As a result, it was found that the predicted value by the forecast combination method was closer to the actual value than the predicted values of the machine learning techniques. In addition, the forecast combination method showed stable prediction results unlike volatile predicted values by machine learning techniques.

• Informatization Policy
• /
• v.29 no.4
• /
• pp.26-42
• /
• 2022

In order to evaluate the usefulness of metaverse learning from the learner's point of view, this study 1) evaluated whether the expectancy-value of the class was satisfied before and after the learner used the metaverse learning platform and 2) verified factors affecting metaverse learning satisfaction with regard to the self-efficacy and expectancy-value of learners. Expectancy-value was evaluated by the learning effect, communication, class involvement, and learning attitude, whereas self-efficacy was evaluated by preference for task difficulty, self-regulation efficacy, and self-confidence. As a result of a study targeting 70 college students who applied for a few courses using the metaverse platform at a university in the northeastern part of Seoul, learners were found to have high expectations and values for learning before using the metaverse platform, but both were not statistically satisfied after use. In addition, the higher the self-efficacy of the learner, the higher the satisfaction with the metaverse learning, and statistically significant results were found in the task-difficulty preference and self-regulatory efficacy among the sub-factors of self-efficacy. There is a negative causal relationship between expectancy-value factors and satisfaction with metaverse learning. This study implies that it is a learner-centered evaluation of metaverse learning, revealing the expectancy-value effect and factors influencing the satisfaction with metaverse learning.

• The Mathematical Education
• /
• v.62 no.3
• /
• pp.327-340
• /
• 2023

The value that students consider important in math learning may vary depending on the student's socio-cultural background and personal experience. Although socio-cultural backgrounds are very diverse, I considered overseas vs domestic Koreans, and secondary school levels as variables in terms of students' educational experiences. Overseas students had a lower perception of the value in mathematics than domestic students, especially about understanding mathematics knowledge and the value of the latest teaching and learning methods. Middle school students perceived the value of mathematics as an activity higher than that of high school students, and high school students perceived student agency as a higher value than middle school students. In addition, I considered meta-affect as one of the individual students' experiences, finally meta-affect was a variable that could explain value perception in math learning, and in particular, affective awareness of achievement, affective evaluation of value, and affective using were significant. From the results, I suggested that research on ways to improve the value and the meta-affect in math learning, test to measure the value of students in math learning, the expansion of research subjects to investigate the value in math learning, and a teacher who teaches overseas Koreans are needed.

• Journal of The Korean Society of Integrative Medicine
• /
• v.2 no.2
• /
• pp.1-12
• /
• 2014

Purpose : The purpose of this study was to investigate the effect that the academic achievement of the students about the evidence based learning investigates the learning utility value about and the request. Method : The agreement of college students explaining the purpose of research for 12 weeks against 17 students and investigate through a questionnaire. The level of academic achievement according to the sex and claim showed a characteristic with a percentage. An utility investigate the descriptive epidemiologic characteristic about the class of the evidence based learning. Result : The most of college students the level of academic achievement and requests the expected grade of the students about the evidence based learning wanted the 'high' grade of 9 persons, 'middle' grade of 8 persons in the part and the expectation for the class taken so much was high(p<.05). There was the significant different in the utility aspect in the need of the evidence based learning, homework solution, learning synergy effect improvement, and reference search ability improvement(p<.05). Conclusion : These finding revealed that the evidence based learning the satisfaction with class raises the improvement and utility value, and provided the need and the has to develop the educational model which the college students contentment raises an improvement after this opportunity for the new recognition.

• Proceedings of the Korea Society for Simulation Conference
• /
• 2001.10a
• /
• pp.37-43
• /
• 2001

New product development has been a key element fur organizational evolution. The bulk of research about new product strategy has focused solely on new product development function itself. This paper investigates cross-functional elements in new product development. More specifically, we suggest that there must exist a fit between new product strategy and value chain strategy. It means that, in order to support new product development activity, there must exist a relevant value chain strategy. We consider three types of integration - internal integration, customer integration, and supplier integration - as strategic elements of value chain strategy. For the case of new product strategy, we consider market newness and product technology unfamiliarity as strategic elements. We also consider two types of learning characteristic, i.e., \\\"fast-adaptive learning\\\" and \\\"slow-adaptive leaning\\\" as control factor. Learning characteristic represents firms organizational capability related with organizational learning. For example, fur fast-adaptive learning case, the effect of integration appears early in time. System dynamics simulation is employed to verify our research framework. The results exhibit that there must exist cross-functional relationships between value chain strategy and new product strategy in order to shorten total development time.al development time.

• Journal of Venture Innovation
• /
• v.4 no.1
• /
• pp.23-38
• /
• 2021

This paper will examine the service quality of online education on service value, learning immersion, major satisfaction. The result structural equation model show that service quality of online education had a positive impact on service value. First, system convenience, system quality, interaction had a positive impact on service value. Second, System convenience, system quality, interaction had a positive impact on learning immersion Third, Online education on service value had a positive impact on Major Satisfaction. Fourth, Learning immersion had a positive impact on Major Satisfaction.

• The Journal of Information Systems
• /
• v.32 no.1
• /
• pp.179-199
• /
• 2023

Purpose This study builds a prediction model to find stocks that can reach intrinsic value among KOSPI and KOSDAQ-listed companies to improve the stability and profitability of the stock investment. And investment simulations are conducted to verify whether stock investment performance is improved by comparing the prediction model, random stock selection, and the market indexes. Design/methodology/approach Value investment theory and machine learning techniques are applied to build the model. Various experiments find conditions such as the algorithm with the best predictive performance, learning period, and intrinsic value-reaching period. This study selects stocks through the prediction model learned with inventive variables, does not limit the holding period after buying to reach the intrinsic value of the stocks, and targets all KOSPI and KOSDAQ companies. The stock and financial data are collected for 21 years (2001-2021). Findings As a result of the experiment, using the random forest technique, the prediction model's performance was the best with one year of learning period and within one year of the intrinsic value reaching period. As a result of the investment simulation, the cumulative return of the prediction model was up to 1.68 times higher than the random stock selection and 17 times higher than the KOSPI index. The usefulness of the prediction model was confirmed in that the number of intrinsic values reaching the predicted stock was up to 70% higher than the random selection.

• Journal of IKEEE
• /
• v.23 no.4
• /
• pp.1187-1194
• /
• 2019

This paper proposes a CNN architecture as value function network of an artificial intelligence Othello game agent and its learning scheme using reinforcement learning algorithm. We propose an approach to construct the value function network by using CNN to learn the records of professional players' real game and an approach to enhance the network parameter by learning from self-play using reinforcement learning algorithm. The performance of value function network CNN was compared with existing ANN by letting two agents using each network to play games each other. As a result, the winning rate of the CNN agent was 69.7% and 72.1% as black and white, respectively. In addition, as a result of applying the reinforcement learning, the performance of the agent was improved by showing 100% and 78% winning rate, respectively, compared with the network-based agent without the reinforcement learning.