• 제어로봇시스템학회:학술대회논문집
• /
• 2003.10a
• /
• pp.1300-1304
• /
• 2003

The following paper deals with a control problem of a mobile robot in home network environment. The home network causes the mobile robot to communicate with sensors to get the sensor measurements and to be adapted to the environment changes. To get the improved performance of control of a mobile robot in spite of the change in home network environment, we use the fuzzy inference system with multiple reward reinforcement learning. The multiple reward reinforcement learning enables the mobile robot to consider the multiple control objectives and adapt itself to the change in home network environment. Multiple reward fuzzy Q-learning method is proposed for the multiple reward reinforcement learning. Multiple Q-values are considered and max-min optimization is applied to get the improved fuzzy rule. To show the effectiveness of the proposed method, some simulation results are given, which are performed in home network environment, i.e., LAN, wireless LAN, etc.

• Journal of Korea Multimedia Society
• /
• v.25 no.1
• /
• pp.52-60
• /
• 2022

Few-shot learning is a sub-area of machine learning problems, which aims to classify target images that only contain a few labeled samples for training. As a representative few-shot learning method, the Prototypical network has been received much attention due to its simplicity and promising results. However, the Prototypical network uses the sample mean of samples from the same class as the prototypes of that class, which easily results in learning uncharacteristic features in the low-data scenery. In this study, we propose to use local descriptors (i.e., patches along the channel within feature maps) from the same class to explicitly obtain more representative prototypes for Prototypical Network so that significant intra-class feature information can be maintained and thus improving the classification performance on few-shot learning tasks. Experimental results on various benchmark datasets including mini-ImageNet, CUB-200-2011, and tiered-ImageNet show that the proposed method can learn more discriminative intra-class features by the local descriptors and obtain more generic prototype representations under the few-shot setting.

• Archives of design research
• /
• v.18 no.2 s.60
• /
• pp.255-264
• /
• 2005

The development of personal computers and the Internet led to changes of instructor-student and student-student relationships in design education. New resources of design education with proper materials and tools have to be offered for teaching and learning. The purpose of this research is to develop Basic Graphic Tools(BGTs) on screen that help students to practice the design principles, and to evaluate the BGTs' usability to enhance the credibility of extra tools for online design learning. The BGTs has to be developed to minimize the gap between the online and traditional classroom settings and to maximize the diverse advantage of lecture content driven from technology. Students were given certain design requirements to explore BGTs and a learning guide/performance test after the lecture to evaluate the tools. The test was performed under the e-learning program serviced by C college. This research has presented some encouraging results and raises questions toward future work: how do BGTs impact the feasibility of an online design lecture.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.05a
• /
• pp.538-540
• /
• 2021

In this paper, we proposed a smart target detection system that detects and recognizes a designated target to provide relative motion information when performing a target detection mission of a drone. The proposed system focused on developing an algorithm that can secure adequate accuracy (i.e. mAP, IoU) and high real-time at the same time. The proposed system showed an accuracy of close to 1.0 after 100k learning of the Google Inception V2 deep learning model, and the inference speed was about 60-80[Hz] when using a high-performance laptop based on the real-time performance Nvidia GTX 2070 Max-Q. The proposed smart target detection system will be operated like a drone and will be helpful in successfully performing surveillance and reconnaissance missions by automatically recognizing the target using computer image processing and following the target.

• Journal of Korea Society of Digital Industry and Information Management
• /
• v.10 no.3
• /
• pp.61-70
• /
• 2014

The government supports the various systems on the premise of life-long education for improving the qualification of its people and supporting the lifelong career system. These systems help the educational activities for every generation from the teens. Each government department supports the learning activities for those who are related. In addition, each government department systemizes the regular education for advancing the performance of civil servant, by obliging a self-directed system to their member as well as by reflecting this educational activities in a promotion. This style of learning is provided not only offline but also online. However, not all necessary educational contents is furnished within the department. It is sometimes necessary to cooperate with the commissioned education and a professional educational institution. This thesis suggests three model systems while comparing one another. This paper suggests three system model. The first model is e-contents open place. The second model is e-contents open place. Finally, the suggest model is mixed form.

• Journal of Information Technology Services
• /
• v.20 no.5
• /
• pp.147-158
• /
• 2021

The digital transformation and COVID-19 are also causing major changes in teaching-learning methods. The biggest change is the spread of remote training and the emergence of various innovative learning platforms. Distance education has been criticized for not meeting technology trends and field demands..However, the problem of distance education is being solved through a system that supports various interactions and collaborations and supports customized learning paths. The researcher conducted a case study on domestic and foreign learning platforms that provide non-face-to-face ICT education. Based on the case study results, the researcher presented the functional characteristics of a learning platform that effectively supports non-face-to-face learning. In common, these sites faithfully supported the basic functions of the information system. In addition to learning progress check and learning guidance, some innovative learning platforms were providing differentiated functions in practice support, performance management, mentoring, learning data analysis, curation provision, and CDP support. Most learning platforms supported one-way, superficial interaction. If the platform effectively supports a variety of learning experiences and provides an integrated learning experience thanks to the development of IT technology, user satisfaction with the learning platform, intention to continue learning, and achievement will increase.

• Journal of the Korea Convergence Society
• /
• v.13 no.2
• /
• pp.303-312
• /
• 2022

In this study, the importance-performance analysis (IPA) of the content components of remote graduate students was conducted. To this end, an online survey of 221 remote graduate students at J University obtained the following results. First, the importance of content components by area was in the order of learning content, interaction, teaching-learning strategy, and evaluation, and the degree of execution was in the order of teaching-learning strategy, interaction, evaluation, and learning content. Second, it was found that there were significant differences between importance and performance in the four areas of content components: learning content, teaching-learning strategy, interaction, and evaluation. The importance-execution analysis (IPA) was conducted in two dimensions: region-specific and item-specific, and the results are as follows. The learning content was found to be the maintenance area, the teaching-learning strategy and interaction were the key improvement areas, and the evaluation area was the overinvestment area. The results of this study can be used as basic data to diagnose the present of remote graduate school content and to gauge what needs to be improved in the future based on it.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.10 no.6
• /
• pp.2527-2545
• /
• 2016

Measuring the similarity of given samples is a key problem of recognition, clustering, retrieval and related applications. A number of works, e.g. kernel method and metric learning, have been contributed to this problem. The challenge of similarity learning is to find a similarity robust to intra-class variance and simultaneously selective to inter-class characteristic. We observed that, the similarity measure can be improved if the data distribution and hidden semantic information are exploited in a more sophisticated way. In this paper, we propose a similarity learning approach for retrieval and recognition. The approach, termed as LDA-FEK, derives free energy kernel (FEK) from Latent Dirichlet Allocation (LDA). First, it trains LDA and constructs kernel using the parameters and variables of the trained model. Then, the unknown kernel parameters are learned by a discriminative learning approach. The main contributions of the proposed method are twofold: (1) the method is computationally efficient and scalable since the parameters in kernel are determined in a staged way; (2) the method exploits data distribution and semantic level hidden information by means of LDA. To evaluate the performance of LDA-FEK, we apply it for image retrieval over two data sets and for text categorization on four popular data sets. The results show the competitive performance of our method.

• Journal of Korean Association for Spatial Structures
• /
• v.21 no.4
• /
• pp.73-80
• /
• 2021

A mid-story isolation system was proposed for seismic response reduction of high-rise buildings and presented good control performance. Control performance of a mid-story isolation system was enhanced by introducing semi-active control devices into isolation systems. Seismic response reduction capacity of a semi-active mid-story isolation system mainly depends on effect of control algorithm. AI(Artificial Intelligence)-based control algorithm was developed for control of a semi-active mid-story isolation system in this study. For this research, an practical structure of Shiodome Sumitomo building in Japan which has a mid-story isolation system was used as an example structure. An MR (magnetorheological) damper was used to make a semi-active mid-story isolation system in example model. In numerical simulation, seismic response prediction model was generated by one of supervised learning model, i.e. an RNN (Recurrent Neural Network). Deep Q-network (DQN) out of reinforcement learning algorithms was employed to develop control algorithm The numerical simulation results presented that the DQN algorithm can effectively control a semi-active mid-story isolation system resulting in successful reduction of seismic responses.

• The Journal of the Korea Contents Association
• /
• v.10 no.8
• /
• pp.451-460
• /
• 2010

The purpose of this paper is to present the effective operation method of cyber home learning system 2.0. Cyber home learning system 2.0 which plays an important role to produce and share data of internet users establishing user-centered internet environment is recently introduced as web 2.0 is beginning to appear. This research is based on the case of the operation on cyber home learning system, research literature of web 2.0 as well as expert consulting and performance test. After doing this research, cyber home learning system 2.0 can bring enormous educational effect when it has online education using web-based discussion tool and long distance consulting linking with off line school education. Redundant system configuration, the distributed processing also may be provided by the service was smooth. Moreover, it demands monitoring team consisting of student, parents and teacher in order to provide developmental service.