• Korean Journal of Cognitive Science
• /
• v.28 no.2
• /
• pp.111-131
• /
• 2017

The connectionist model is one approach to studying language processing from a computational perspective. And building a representation in the connectionist model study is just as important as making the structure of the model in that it determines the level of learning and performance of the model. The connectionist model has been constructed in two different ways: localist representation and distributed representation. However, the localist representation used in the previous studies had limitations in that the unit of the output layer having a rare target activation value is inactivated, and the past distributed representation has the limitation of difficulty in confirming the result by the opacity of the displayed information. This has been a limitation of the overall connection model study. In this paper, we present a new method to induce distributed representation with local representation using abstraction of information, which is a feature of restricted Boltzmann machine, with respect to the limitation of such representation of the past. As a result, our proposed method effectively solves the problem of conventional representation by using the method of information compression and inverse transformation of distributed representation into local representation.

• The Journal of the Korea Contents Association
• /
• v.11 no.8
• /
• pp.123-133
• /
• 2011

In this paper, we propose a new method for toon shading using 3D texture which renders 3d objects in a cartoon style. The conventional toon shading using 1D texture displays shading tone by computing the relative position and orientation between a light vector and surface normal. The 1D texture alone has limits to express the various tone change according to any viewing condition. Therefore Barla et. al. replaces a 1D texture with a 2D texture whose the second dimension corresponds to the view-dependent effects such as level-of-abstraction, depthof-field. The proposed scheme extends 2D texture to 3D texture by adding one dimension with the geometric information of 3D objects such as curvature, saliency, and coordinates. This approach supports two kinds of extensions for cartoon style diversification. First, we support "shape exaggeration effect" to emphasize silhouette or highlight according to the geometric information of 3D objects. Second, we further incorporate "cartoon specific effect", which is examples of screen tone and out focusing frequently appeared in cartoons. We demonstrate the effectiveness of our approach through examples that include a number of 3d objects rendered in various cartoon style.

• Korean Journal of Biological Psychiatry
• /
• v.8 no.2
• /
• pp.258-265
• /
• 2001

Objectives : Chronic alcohol consumption has been known to result in various neurocognitive deficits. Many neuropsychological studies revealed that the major disturbances occurred in the executive function, learning and short-term memory, visuospatial performance function, perceptuo-motor skills, and abstraction and problem solving abilities. This study was done to identify which cognitive areas might be mainly affected. Methods : The cognitive disturbance was evaluated using the Korean Version of the Mini Mental State Examination(MMSEK) and the 7 Minute Screen(7MS) in male inpatients with alcohol dependence(N=3 : as well as in age and education level matched healthy male controls(N=30). Four individual tests of the 7MS were consisted of the Benton Temporal Orientation Test, the Enhanced Cued Recall, the Clock Drawing and the Category Fluency. Results : 1) The average scores of four individual test of the 7MS for the alcoholics were $2.77{\pm}4.38$ for the Benton Temporal Orientation Test, $13.90{\pm}2.02$ for the Memory Test(the Cued Recall $6.77{\pm}1.94$, the Uncued Recall $7.10{\pm}2.45$), $5.84{\pm}1.86$ for the Clock Drawing, and $12.58{\pm}3.29$ for the Category Fluency. Except the Benton Temporal Orientation Test, there were statistically significant differences between test scores of alcoholics and those of controls(p<0.01). 2) The alcoholics who had MMSE-K score <24 were 9.68%. The average(${\pm}S.D.$) score of the MMSE-K for the patient group($27.23{\pm}2.62$) was significantly(p<0.001) lower than that of the healthy controls($29.20{\pm}1.24$). There were no statistically significant differences between four individual test scores of the 7MS of alcoholics with the MMSE-K score <24(N=3) and those of alcoholics with the MMSE-K score ${\geq}24$(N=28). 3) Four individual test scores of the 7MS seemed to have statistically significant association with such variables as MMSE-K, duration of alcohol drinking, blood magnesium concentration, liver function and thyroid function. Conclusion : Mild deficits of cognitive areas such as orientation, memory, visuospatial abilities and verbal fluency could be found in alcohol dependence.

• Cartoon and Animation Studies
• /
• s.49
• /
• pp.697-712
• /
• 2017

In recent years, film and animation for cinematic release have embraced stereoscopic vision and the three-dimensional depth it creates for the viewer. The maturation of consumer-level virtual reality (VR) technology simultaneously spurred a wave of media productions set within 3D space, ranging from computer games to pornographic videos, to Academy Award-nominated animated VR short film Pearl. All of these works rely on stereoscopic fusion through stereopsis, that is, the perception of depth produced by the brain from left and right images with the amount of binocular parallax that corresponds to our eyes. They aim to emulate normal human vision. Within more experimental practices however, a fully rendered 3D space might not always be desirable. In my own abstract animation work, I tend to favour 2D flatness and the relative obfuscation of spatial relations it affords, as this underlines the visual abstraction I am pursuing. Not being able to immediately understand what is in front and what is behind can strengthen the desired effects. In 2015, Jeffrey Shaw challenged me to create a stereoscopic work for Animamix Biennale 2015-16, which he co-curated. This prompted me to question how stereoscopy, rather than hyper-defining space within three dimensions, might itself be used to achieve a confusion of spatial perception. And in turn, how abstract and experimental moving image practices can benefit from stereoscopy to open up new visual and narrative opportunities, if used in ways that break with, or go beyond stereoscopic fusion. Noteworthy works which exemplify a range of non-traditional, expanded approaches to binocular vision will be discussed below, followed by a brief introduction of the stereoscopic animation loop III=III which I created for Animamix Biennale. The techniques employed in these works might serve as a toolkit for artists interested in exploring a more experimental, expanded engagement with stereoscopy.

• Journal of KIISE:Software and Applications
• /
• v.29 no.5
• /
• pp.295-310
• /
• 2002

Software architecture representing a common high-level abstraction of a system can be used as a basis for creating mutual understanding among all stakeholders of the system. In determining a software architecture's fitness with respect to its desired qualities as well as in improving a software architecture, software architecture evaluation is importantly performed. However moat of architecture evaluation methods are not still sufficient in that they do not explicitly consider artifacts discussed during architecture evaluation and their processes are net systematic. As a result, we are hard to follow them. To address these problems, this paper presents the method to evaluate systematically a software architecture with respect to its desired qualities. In this approach, the functional and non-functional requirements are separately handled, and software architecture is represented in the 4+1 view model using UML. Through this initial consideration, the important artifacts such as goals, scope, and target of evaluation are clearly determined. Also, the method provides the well defined process to produce the important evaluation artifacts such as sub-designs, design decisions, rationale, qualities from inputs. In addition, it enables us to determine satisfaction of a architecture with respect its desired qualities or improve a architecture through the structured evaluation results.

• Journal of Intelligence and Information Systems
• /
• v.23 no.2
• /
• pp.1-17
• /
• 2017

The deep learning framework is software designed to help develop deep learning models. Some of its important functions include "automatic differentiation" and "utilization of GPU". The list of popular deep learning framework includes Caffe (BVLC) and Theano (University of Montreal). And recently, Microsoft's deep learning framework, Microsoft Cognitive Toolkit, was released as open-source license, following Google's Tensorflow a year earlier. The early deep learning frameworks have been developed mainly for research at universities. Beginning with the inception of Tensorflow, however, it seems that companies such as Microsoft and Facebook have started to join the competition of framework development. Given the trend, Google and other companies are expected to continue investing in the deep learning framework to bring forward the initiative in the artificial intelligence business. From this point of view, we think it is a good time to compare some of deep learning frameworks. So we compare three deep learning frameworks which can be used as a Python library. Those are Google's Tensorflow, Microsoft's CNTK, and Theano which is sort of a predecessor of the preceding two. The most common and important function of deep learning frameworks is the ability to perform automatic differentiation. Basically all the mathematical expressions of deep learning models can be represented as computational graphs, which consist of nodes and edges. Partial derivatives on each edge of a computational graph can then be obtained. With the partial derivatives, we can let software compute differentiation of any node with respect to any variable by utilizing chain rule of Calculus. First of all, the convenience of coding is in the order of CNTK, Tensorflow, and Theano. The criterion is simply based on the lengths of the codes and the learning curve and the ease of coding are not the main concern. According to the criteria, Theano was the most difficult to implement with, and CNTK and Tensorflow were somewhat easier. With Tensorflow, we need to define weight variables and biases explicitly. The reason that CNTK and Tensorflow are easier to implement with is that those frameworks provide us with more abstraction than Theano. We, however, need to mention that low-level coding is not always bad. It gives us flexibility of coding. With the low-level coding such as in Theano, we can implement and test any new deep learning models or any new search methods that we can think of. The assessment of the execution speed of each framework is that there is not meaningful difference. According to the experiment, execution speeds of Theano and Tensorflow are very similar, although the experiment was limited to a CNN model. In the case of CNTK, the experimental environment was not maintained as the same. The code written in CNTK has to be run in PC environment without GPU where codes execute as much as 50 times slower than with GPU. But we concluded that the difference of execution speed was within the range of variation caused by the different hardware setup. In this study, we compared three types of deep learning framework: Theano, Tensorflow, and CNTK. According to Wikipedia, there are 12 available deep learning frameworks. And 15 different attributes differentiate each framework. Some of the important attributes would include interface language (Python, C ++, Java, etc.) and the availability of libraries on various deep learning models such as CNN, RNN, DBN, and etc. And if a user implements a large scale deep learning model, it will also be important to support multiple GPU or multiple servers. Also, if you are learning the deep learning model, it would also be important if there are enough examples and references.

• Journal of the Korean earth science society
• /
• v.25 no.5
• /
• pp.336-347
• /
• 2004

This study investigated the effects of the application of a variety of ICTs cause the effects on self-directed learning capability, creativity and problem-solving ability. In order to achieve the above aim, Web-Based Instructions(WBI) and instructions using CD-ROM Titles for the unit of 'the solar system and the galaxy' were applicated and analyzed which belongs to the area of 'the earth' in the subject 'science' for high school students. Instructions using WBI materials and CD-ROM titles were found to be effective on 'self-conception', 'creativity', 'future inclination', 'self-assessment ability', 'openness' and' initiative' improvement all of which belong to self-directed learning characteristics. They did not, however, show meaningful effect on improving 'learning eagemess' and 'responsibility' improvement. On looking into self-directed learning characteristics according to prerequisite learning levels, both groups and these for instruction using CD-ROM learning materials were found to have no effect on interaction. With respect to problem-solving ability improvement which is characteristic of the instruction using ICTs, WBI proved more fruitful than instruction using CD-ROM titles on improving scholastic achievement level. WBI was effective on 'fluency', 'originality' and 'resistance to premature closure'. It on the other hand, was of no use on 'abstraction of titles' and 'elaborateness' These results came from the following characteristics: WBI came into effect on 'fluency' and 'originality' in the areas of variety and vitality, which are characteristic of WBI. In the area of resistance to premature closure WBI was effective on organizing learning contents owing to the animation of picture materials which are variously presented in the web site. As a result of WBI questionnaire about WBI, an excellent effect on the structure of display, quantity of information, indication and instruction, supplementary study and further study were discussed.

• The Journal of the Convergence on Culture Technology
• /
• v.7 no.4
• /
• pp.273-280
• /
• 2021

GangAm Song Sungyong (1913~1999) was from Gimje, Jeollabuk-do. And he is a calligrapher from the last Confucian scholar of the 20th century and a writer of poetry, calligraphy, and painting. While wearing a topcoat and hanbok for the rest of his life, he approached the study, thought, and calligraphy art of Yoo Jae Song Kimyeon and Kojae Lee Byungeun from the perspective of 'GucheSinyong'. And he kept the philosophical subjectivity that tries to maintain character and a right mind. It was based on 'Guche', which is a faithful succession to the reverence of the old. When I was young, I practiced Mibul and Dong Kichang with Gu Yangsun as the center, and Hwang Jeonggyeon, Hanye, Oh Heejae, and Sojeon typefaces were grafted together during the national exhibition. Then, in 1965 (age 53), he moved to Jeonju, and learned several typefaces such as Chusa typeface on his own, creating a Gangam typeface without any obstacles. And he created 'Sinyong' with Windy Bamboo painting, which embodied strange and unconventional meanings. In addition, he re-established his identity by reexamining the fundamental spirit and natural aesthetics of calligraphy, and based on this, he greatly contributed to expanding the aesthetics of modern calligraphy and painting art by pursuing an aesthetic that explores novelty. In particular, Windy Bamboo painting has strong abstraction based on the principle of 'drawing the will'. And, in terms of discipline, the ethical aesthetic of Express Tao with pictures (畵以載道), which expresses the high level of elegance of observing small things in a big way, based on deep research on the logic of things and fulfilling human nature. By implementing it, Tao and Art become one. The Jeonbuk calligraphy group achieved the greatest prosperity in the Gangam era, and at the end of the 20th century, it entered a period of revival as it established itself as the central calligraphy group of Korean calligraphy.