• Proceedings of the Korean Information Science Society Conference
• /
• 2004.04b
• /
• pp.622-624
• /
• 2004

강화 학습(Reinforcement Learning)을 실제 문제에 적용하는 데 있어 가장 큰 문제는 차원성의 저주(Curse of dimensionality)였다 문제가 커짐에 따라 목적을 이루기 위해서 더 많은 단계의 판단이 필요하고 이에 따라 문제의 해결이 지수적으로 어려워지게 된다. 이를 해결하기 위해 문제를 여러 단계로 나누어 단계별로 학습하는 계층적 강화 학습(Hierarchical Reinforcement Learning)이 제시된 바 있다 하지만 대부분의 계층적 강화 학습 방법들은 사전에 문제의 구조를 아는 것을 전제로 하며 큰 사이즈의 문제를 간단히 표현할 방법을 제시하지 않는다. 따라서 이들 방법들도 실제적인 문제에 바로 적용하기에는 적합하지 않다. 최근 이루어진 복잡계 네트워크(Complex Network)에 대한 연구에 착안하여 본 논문은 자기조직화하는 생장 네트워크(Self organizing growing network)를 기반으로 한 간단한 환경 표현 모델을 사용하는 강화 학습 알고리즘을 제안한다 네트웍은 복잡계 네트웍이 갖는 성질들을 유지하도록 자기 조직화되고, 노드들 간의 거리는 작은 세상 성질(Small World Property)에 따라 전체 네트웍의 큰 사이즈에 비해 짧게 유지된다. 즉 판단해야할 단계의 수가 적게 유지되기 때문에 이 방법으로 차원성의 저주를 피할 수 있다.

• Journal of Korea Multimedia Society
• /
• v.24 no.10
• /
• pp.1391-1402
• /
• 2021

This paper focuses use context-sensitive spelling error correction using generative adversarial network. Generative adversarial network[1] are attracting attention as they solve data generation problems that have been a challenge in the field of deep learning. In this paper, sentences are generated using word embedding information and reflected in word distribution representation. We experiment with DCGAN[2] used for the stability of learning in the existing image processing and D2GAN[3] with double discriminator. In this paper, we experimented with how the composition of generative adversarial networks and the change of learning corpus influence the context-sensitive spelling error correction In the experiment, we correction the generated word embedding information and compare the performance with the actual word embedding information.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.13 no.4
• /
• pp.2180-2197
• /
• 2019

With the proliferation of the Android malicious applications, malware becomes more capable of hiding or confusing its malicious intent through the use of code obfuscation, which has significantly weaken the effectiveness of the conventional defense mechanisms. Therefore, in order to effectively detect unknown malicious applications on the Android platform, we propose DroidVecDeep, an Android malware detection method using deep learning technique. First, we extract various features and rank them using Mean Decrease Impurity. Second, we transform the features into compact vectors based on word2vec. Finally, we train the classifier based on deep learning model. A comprehensive experimental study on a real sample collection was performed to compare various malware detection approaches. Experimental results demonstrate that the proposed method outperforms other Android malware detection techniques.

• ETRI Journal
• /
• v.45 no.1
• /
• pp.93-104
• /
• 2023

Coreference resolution is a task in discourse analysis that links several headwords used in any document object. We suggest pointer networks-based coreference resolution for Korean using multi-task learning (MTL) with an attention mechanism for a hierarchical structure. As Korean is a head-final language, the head can easily be found. Our model learns the distribution by referring to the same entity position and utilizes a pointer network to conduct coreference resolution depending on the input headword. As the input is a document, the input sequence is very long. Thus, the core idea is to learn the word- and sentence-level distributions in parallel with MTL, while using a shared representation to address the long sequence problem. The suggested technique is used to generate word representations for Korean based on contextual information using pre-trained language models for Korean. In the same experimental conditions, our model performed roughly 1.8% better on CoNLL F1 than previous research without hierarchical structure.

• Educational Technology International
• /
• v.9 no.2
• /
• pp.79-95
• /
• 2008

This paper aims to investigate modeling scripts based on the mode of group interaction in a computer-supported collaborative learning environment. Based on a literature review, this paper assumes that group interaction and its mode would have strong influence on the online collaborative learning process, and furthermore lead learners to create and share significant knowledge within a group. This paper deals with two different modes of group interaction- distributed and shared interaction. Distributed interaction depends on the external representation of individual knowledge, while shared interaction is concerned with sharing knowledge in group action. In order to facilitate these group interactions, this paper emphasizes the utilization of appropriate CSCL scripts, and then proposes the conceptual framework of CSCL scripts which integrate the existing scripts such as implicit, explicit, internal and external scripts. By means of the model regarding CSCL scripts based on the mode of group interaction, the implications for research on the design of CSCL scripts are explored.

• Journal of the Korean association of regional geographers
• /
• v.18 no.4
• /
• pp.445-463
• /
• 2012

This paper focuses on media literacy as a trial for reestablishing the relationship between media and geography education. So far, a geographical phenomenon represented through media has been treated as a transparent window on the world, but now needs to be recognized as a product of representation constructed socially by a range of subjects and their purposes. The epistemological turn of media has brought interest on social construction and media literacy in terms of teaching and learning. It is required that teaching and learning geography through media should be turned from the existing massmedia in education(or the education using media) to the education for fostering an active media literacy to analyze and reason critically how the media as text is constructed and selected. This geography education as media literacy is very important because it enables students to reveal the ideology and power relation embedded in the media as text, as well as to stimulate and enrich their geography imagination through an active work.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.22 no.3
• /
• pp.319-327
• /
• 2012

In this paper, we present a sign language recognition model which does not use any wearable devices for object tracking. The system design issues and implementation issues such as data representation, feature extraction and pattern classification methods are discussed. The proposed data representation method for sign language patterns is robust for spatio-temporal variances of feature points. We present a feature extraction technique which can improve the computation speed by reducing the amount of feature data. A neural network model which is capable of incremental learning is described and the behaviors and learning algorithm of the model are introduced. We have defined a measure which reflects the relevance between the feature values and the pattern classes. The measure makes it possible to select more effective features without any degradation of performance. Through the experiments using six types of sign language patterns, the proposed model is evaluated empirically.

• Conservation Science in Museum
• /
• v.31
• /
• pp.55-70
• /
• 2024

Style transfer algorithms are currently undergoing active research and are used, for example, to convert ordinary images into classical painting styles. However, such algorithms have yet to produce appropriate results when applied to Korean cultural heritage images, while the number of cases for such applications also remains insufficient. Accordingly, this study attempts to develop a style transfer algorithm that can be applied to styles found among Korean cultural heritage. The algorithm was produced by improving data comprehension by enabling it to learn meaningful characteristics of the styles through representation learning and to separate the cultural heritage from the background in the target images, allowing it to extract the style-relevant areas with the desired color and texture from the style images. This study confirmed that, by doing so, a new image can be created by effectively transferring the characteristics of the style image while maintaining the form of the target image, which thereby enables the transfer of a variety of cultural heritage styles.

• 한국게임학회지
• /
• v.8 no.1
• /
• pp.32-38
• /
• 2011

The help or guide sentences of educational math games which use mathematical statements need to represent graphical forms for the learners of the game generation whose cognitive style is graphic first. In this paper, we proposed an object-based visual representation method for mathematical statements. It has object-based description rules to use graphical symbols and mathematical symbols with text words. It is easy to describe or to understand accurately mathematical meaning and is also fast for learners to read for understanding. The proposed method is good for learners of the game generation to get the help as scaffolding for learning math by educational games.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.1
• /
• pp.402-410
• /
• 2017

For many years, researchers in face description area have been representing and recognizing faces based on different methods that include subspace discriminant analysis, statistical learning and non-statistics based approach etc. But still automatic face recognition remains an interesting but challenging problem. This paper presents a novel and efficient face image representation method based on Optimized Local Ternary Pattern (OLTP) texture features. The face image is divided into several regions from which the OLTP texture feature distributions are extracted and concatenated into a feature vector that can act as face descriptor. The recognition is performed using nearest neighbor classification method with Chi-square distance as a similarity measure. Extensive experimental results on Yale B, ORL and AR face databases show that OLTP consistently performs much better than other well recognized texture models for face recognition.