• KIPS Transactions on Software and Data Engineering
• /
• v.11 no.8
• /
• pp.331-338
• /
• 2022

Successor representation (SR) is a model of human reinforcement learning (RL) mimicking the underlying mechanism of hippocampal cells constructing cognitive maps. SR utilizes these learned features to adaptively respond to the frequent reward changes. In this paper, we evaluated the performance of SR under the context where changes in latent variables of environments trigger the reward structure changes. For a benchmark test, we adopted SR-Dyna, an integration of SR into goal-driven Dyna RL algorithm in the 2-stage Markov Decision Task (MDT) in which we can intentionally manipulate the latent variables - state transition uncertainty and goal-condition. To precisely investigate the characteristics of SR, we conducted the experiments while controlling each latent variable that affects the changes in reward structure. Evaluation results showed that SR-Dyna could learn to respond to the reward changes in relation to the changes in latent variables, but could not learn rapidly in that situation. This brings about the necessity to build more robust RL models that can rapidly learn to respond to the frequent changes in the environment in which latent variables and reward structure change at the same time.

• Journal of Energy Engineering
• /
• v.21 no.1
• /
• pp.33-42
• /
• 2012

In this study, long-term greenhouse gas reductions expected passenger sector was used for the MESSAGE. Green Car road map proposed BAU scenario, Enhanced diffusion green car scenario, and price 1, 2 scenarios was configured with four scenarios. Enhanced diffusion green car in the scenario, in 2050 compared to BAU scenario 13% of the emissions will decrease. Price 1 and Price 2 scenario is emissions reduction of 14% compared to BAU. This study consists of six chapters. Introduction of MESSAGE, creation and RES in the year and the target year set a different base line and the passenger building materials sector activities, steps for passenger sector scenario and Based on the results of running the emissions reductions were to describe.

• Journal of the Economic Geographical Society of Korea
• /
• v.6 no.1
• /
• pp.61-77
• /
• 2003

Finns are central economic agents that play an important role in systems of innovation as they take responsibility for generating and diffusing knowledge in both organizational and societal context. They must be considered as learning organizations which interact with other finns and institutions that share their environment. The systems of innovation literature accentuates institutional conditions that influence innovation in sectoral, regional or national levels. Meanwhile, it tends to ignore the complex dimensions of finn practices in relation to learning and innovation activities. In this context, this paper attempts to examine what finns do for sustaining innovation and how they learn to innovate. This is not just critical to know individual finns innovativeness which depends on interactions with environments within and outside the organizational boundary but also to evaluate the regional innovation system potential. In short, it is important to see that finns would attempt to take advantage of distributed knowledge within and across the boundaries of the finn without sticking to particular regional innovation systems. I argue that the more finns of a cluster attempt not only to combine localized sources of knowledge and external sources of knowledge but also to become a learning organization, the more increased regional innovation system potentials can be.

• Journal of the Korean earth science society
• /
• v.29 no.6
• /
• pp.506-519
• /
• 2008

This article explores the potential learning materials and methods of science practice from exhibits, and how those are presented in natural history museums as a feasible science inquiry community. The idea of science inquiry community was offered as a form of science practice that ended with science learning. A grasp of 'scientific practice to learning' is understood as a way to conceive scientific methods as well as facts and understanding knowledge. To get educational implications on the scientific practice of 'earthquake' as a socioscientific topic in the communities, we analyzed 1) the relationship between earth science curriculum and exhibits related to 'earthquake', 2) the educational goals and intentions of educators, and 3) the characteristics of the exhibits in the American Museum of Natural History and in the Smithsonian National Museum of Natural History. The results of this study showed that those museums presented the exhibits consisting of various and practical cases and events of 'earthquakes' as a socioscientific topic related to their curriculum. At the target museum, it was clearly stated that the pursuing educational goals focused on relations with local interests and socioscientific issues. For making earthquakes relevant to visitors, delivering lived experiences with raw data and interactive media was emphasized in exhibit characteristics.

• The Journal of the Acoustical Society of Korea
• /
• v.41 no.2
• /
• pp.199-209
• /
• 2022

Typical anomaly detection algorithms are trained by using prior data. Thus the batch learning based algorithms cause inevitable performance degradation when characteristics of newly incoming normal data change over time. We propose an online anomaly detection algorithm which can consider the gradual characteristic changes of incoming normal data. The proposed algorithm based on one-class classification model includes both offline and online learning procedures. In offline learning procedure, the algorithm learns the prior data to be close to centroid of the latent space and then updates the centroid of the latent space incrementally by new incoming data. In the online learning, the algorithm continues learning by using the updated centroid. Through experiments using public underwater acoustic data, the proposed online anomaly detection algorithm takes only approximately 2 % additional learning time for the incremental centroid update and learning. Nevertheless, the proposed algorithm shows 19.10 % improvement in Area Under the receiver operating characteristic Curve (AUC) performance compared to the offline learning model when new incoming normal data comes.

• The Mathematical Education
• /
• v.61 no.4
• /
• pp.613-629
• /
• 2022

This study explored the heterogeneous latent group on the influence of the learner's perceived math teaching method(instructor-centered, learner-centered) on math self-efficacy. In order to profile the characteristics of the detected latent group, the distribution of variables was confirmed, and multi-group analysis was conducted by SEM. According to the analysis results, two latent groups were detected, and the instructor-type group and the learner-type group were named. As a result of post-hoc analysis, the perception of instructor-centered classes and learner-centered classes, and the perception of math teaching ability were similar between the instructor-type and the learner-type group. But the instructor-type group had higher math self-efficacy, math interest, and math class engagement than the learner-type group. Also, in the instructor-type group, the effect of perception of math teaching ability on math self-efficacy and math class engagement was greater than that of the learner-type group. Whereas, in the learner-type group, the effect of math interest on math self-efficacy and math class engagement was greater than that of the instructor-type group. This study presented a new research method on the influence of math teaching methods on learners by applying the REBUS-PLS method.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.197-205
• /
• 2023

This study is to apply the Havruta teaching-learning method to college students' major classes and analyze the relationship between the effectiveness evaluation of communication skills and sub-factors using a structural equation model. As a result of the study, the communication ability score was different before and after Havruta teaching-learning, and it was found that after Havruta teaching-learning was higher than before Havruta teaching-learning. The path effect was found to be significant in all of the total, direct, and indirect effects among latent variables, except for the relationship between interpretation ability, role-playing ability, and goal-setting ability in the direct effect. In this study, it was found that the Havruta teaching-learning method not only improves creativity and thinking ability, but also improves self-directed learning ability. In addition, it was reconfirmed that it is a teaching-learning method that can develop social skills and communication skills as well as problem-solving skills while experiencing opinions different from one's own. As a result, research on a thorough student-centered teaching-learning method suitable for the Homo Machina era must be continued and its application in the educational field must be implemented.

• Journal of the Korea Society of Computer and Information
• /
• v.13 no.4
• /
• pp.59-69
• /
• 2008

The purpose of this study was to investigate children's Perception of parenting attitude parent and learned helplessness on addictive use of computer game. The data was collected through paper-pencils surveys with 745 students who are attending the 4, 5, and 6 grade of three elementary schools. The instrument used to see how they looked at the parenting attitude of their was Ji-hyun Hwang(2006)'s Perceived Parenting Attitude Inventory, Chang-woo Song(1997)'s Learned Helplessness, and Internet Game Addiction Test by developed Korean Agency Digital Apportunity & Promotion. Statistics and methods used for the data analysis were Cronbach'a alpha, freuency, percentage, Two-Way ANOVA, Pearson's Correlation, and Regression by using SPSS WIN 12.0. The results of this study is described as follows. In this study, prevalence of computer addiction tendency and addiction is 14.3%. Male students showed highest computer addiction game than female students.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.3
• /
• pp.177-187
• /
• 2023

This study aims to validate a tool evaluating MOOCs for higher education from the perspective of education service. Based on the results of related researches, a potential model for evaluating MOOCs (4 factors and 8 sub-factors) was made. An evaluation tool consisting of 18 survey items was delivered to 138 college students. After data cleaning, 136 surveys were used for exploratory factor analysis (principal component analysis. varimax rotation) and reliability analysis that confirmed the fitness of the potential model. Four exploratory constructs and seven sub-factors were extracted: Factor I was labeled as 'Systemic Learning Experience,' Factor II, 'Value Experience,' Factor III, 'Co-creation of Value Experience,' and Factor IV, 'High Order Learning Experience.' Reliability estimates using Cronbach's alpha indicated that the evaluation tool had good internal consistency. In conclusion, the evaluation tool for MOOCs in higher education was proven to be valid and reliable.

• Journal of KIISE:Software and Applications
• /
• v.26 no.8
• /
• pp.1010-1017
• /
• 1999

불완전한 계획 영역 이론은 오류 영역(noisy domain)에서 하나의 상태에 상반된 연산자들이 적용되는 불일치성 문제를 야기할 수 있다. 이 문제를 해결하기 위해서 본 논문은 상태를 기술하기 위해 다치 논리를 도입하여 제어지식으로서의 부정적 선행조건을 학습하는 새로운 방법을 제안한다. 기계에는 알려지지 않은 이러한 제어지식이 인간에게는 반대개념으로 잠재적으로 사용되고 있다. 이러한 잠재된 개념을 학습하기 위해 본 논문은 반대 연산자들로 구성된 사이클을 영역이론으로부터 기계적으로 생성하고, 이 연산자들에 대한 실험을 통해 반대 리터럴(literal)들을 추출한다. 학습된 규칙은 불일치성을 방지하면서 동시에 중복된 선행조건을 제거하여 연산자를 단순화시킬 수 있다.Abstract An incomplete planning domain theory can cause an inconsistency problem in a noisy domain, allowing two opposite operators to be applied to a state. To solve the problem, we present a novel method to learn a negative precondition as control knowledge by introducing a three-valued logic for state description. However, even though the control knowledge is unknown to a machine, it is implicitly known as opposite concept to a human. To learn the implicit concept, we mechanically generate a cycle composed of opposite operators from a domain theory and extract opposite literals through experimenting the operators. A learned rule can simplify the operator by removing a redundant precondition while preventing inconsistency.