• 제어로봇시스템학회:학술대회논문집
• /
• 1991.10a
• /
• pp.866-871
• /
• 1991

The multi-layer neural network possesses the desirable characteristics of parallel distributed processing and learning capacity, by which the uncertain variation of the parameters in the dynamically complex system can be handled adoptively. However the error back propagation algorithm that has been utilized popularly in the learning procedure of the mulfi-Jayer neural network has the significant limitations in the real application because of its slow convergence speed. In this paper, an approach to improve the convergence speed is proposed using the fuzzy logic that can effectively handle the uncertain and fuzzy informations by linguistic level. The effectiveness of the proposed algorithm is demonstrated by computer simulation of PUMA 560 robot manipulator.

• Journal of Korea Multimedia Society
• /
• v.24 no.5
• /
• pp.695-700
• /
• 2021

The Faster RCNN-based shellfish recognition algorithm is introduced for shellfish recognition studies that currently do not have any deep learning-based algorithms in a practical setting. The original feature extraction module is replaced by DenseNet, which fuses multi-level feature data and optimises the NMS algorithm, network depth and merging method; overcoming the omission of shellfish overlap, multiple shellfish and insufficient light, effectively solving the problem of low shellfish classification accuracy. In the complexifier test environment, the test accuracy was improved by nearly 4%. Higher testing accuracy was achieved compared to the original testing algorithm. This provides favourable technical support for future applications of the improved Faster RCNN approach to seafood quality classification.

• Journal of the Korean Mathematical Society
• /
• v.59 no.2
• /
• pp.217-233
• /
• 2022

Although machine learning shows state-of-the-art performance in a variety of fields, it is short a theoretical understanding of how machine learning works. Recently, theoretical approaches are actively being studied, and there are results for one of them, margin and its distribution. In this paper, especially we focused on the role of margin in the perturbations of inputs and parameters. We show a generalization bound for two cases, a linear model for binary classification and neural networks for multi-classification, when the inputs have normal distributed random noises. The additional generalization term caused by random noises is related to margin and exponentially inversely proportional to the noise level for binary classification. And in neural networks, the additional generalization term depends on (input dimension) × (norms of input and weights). For these results, we used the PAC-Bayesian framework. This paper is considering random noises and margin together, and it will be helpful to a better understanding of model sensitivity and the construction of robust generalization.

• Journal of The Korea Institute of Healthcare Architecture
• /
• v.15 no.3
• /
• pp.71-77
• /
• 2009

This study is to understand the situations of special education classroom layout, find differences according to school levels and summarize the findings in order to build up the indicators for special classroom layout. As for elementary school level, special classrooms are using multi-purposes desk or group desk for diverse activities such as basic learning and formation of basic life practice. The most frequent type in classroom layout is Type E which is for diverse coner-learning and play activities and the next is Type C which secures activity space. Because security of dynamic activity which most teachers find problematic is important, it needs more research to secure dynamic activity space within classroom. As for middle school level, the most frequent type in classroom layout is Type B which is equiped for computer aided learning and the next is Type C which secures activity space. Research for systematic layout of activity space is needed in order to secure the spaces of dynamic activity and basic job training. As for high school levels, mostly Type B which emphasizes computer activities is adopted and next is Type F which is capable for job training. The survey about the size of special education classroom proves that most teachers want one and half size classroom which in not such a large classroom. It is expected that more systematic research of special classroom layout according to school levels may reach for rational space layout.

• School Mathematics
• /
• v.15 no.2
• /
• pp.289-314
• /
• 2013

Achievement profile by attribute in Korean students' mathematics was analyzed by applying cognitive diagnostic model, which is the newest measurement theory, to 2011 NAEA(National Assessment of Educational Assessment) results. The results are as follows. As the level of school is higher from 6th grade, 9th grade to 11th grade, the percentage of students mastering cognitive attribute 9(expressions using picture, table, graph, formula, symbol, writing, etc) drastically declined from 78%, 35% to 26%. It is necessary to have learning strategies to reinforce their abilities of expressing table, graph, etc. that higher graders in mathematics are more vulnerable to. Next, the property of mastering cognitive attributes according to gender, multi-cultural family was analyzed. In terms of mathematics, the percentage of girls mastering most of the attribute generally is higher than that of boys from 6th grade to 9th grade, however, boys show higher mastery in almost attributes than girls in the 11th grade. Compared to boys, the part where girls have the most trouble is attribute 9 in mathematics(expressions using picture, table, graph, formula, symbol, writing, etc). As international marriage, influx of foreign workers, etc. increase, the number of students from Korea's multi-cultural families is expected to be higher, therefore, identifying the characteristics of their educational achievement is significant in reinforcing Korea's basic achievement. In mathematics, gap of mastery level of attributes between multi-cultural group and ordinary group is more severe in higher grade and the type of multi-cultural group that needs supports for improving achievement most urgently changed in 6th grade, 9th grade and 11th grade respectively. In the 6th and 11th grade, migrant students from North Korea show the lowest level of mastering attributes, however, in the 9th grade, the mastery rate of immigrant students is lowest. Therefore, there is an implication that supporting plans for improving achievement of students from multi-cultural family should establish other strategies based on the characteristics of school level.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.30 no.6
• /
• pp.1237-1246
• /
• 2020

Recently, several cache side channel attacks have been proposed to extract secret information by exploiting design flaws of the microarchitecture. The Flush+Reload attack, one of the cache side channel attack, can be applied to malicious application attacks due to its properties of high resolution and low noise. In this paper, we proposed a detection system, which detects the cache-based attacks using the PCM(Performance Counter Monitor) for monitoring CPU cache activity. Especially, we observed the variation of each counter value of PCM in case of two kinds of attacks, Spectre attack and secret recovering attack during AES encryption. As a result, we found that four hardware counters were sensitive to cache side channel attacks. Our detector based on machine learning including SVM(Support Vector Machine), RF(Random Forest) and MLP(Multi Level Perceptron) can detect the cache side channel attacks with high detection accuracy.

• Journal of Convergence for Information Technology
• /
• v.10 no.11
• /
• pp.23-29
• /
• 2020

As IoT devices increase exponentially, minimizing MIMO interference and increasing transmission capacity for sending and receiving IoT information through multiple antennas remain the biggest issues. In this paper, secret key-level distribution mechanism using deep learning is proposed to minimize MIMO-based IoT communication noise. The proposed mechanism minimizes resource loss during transmission and reception process by dispersing IoT information sent and received through multiple antennas in batches using deep learning. In addition, the proposed mechanism applied a multidimensional key distribution processing process to maximize capacity through multiple antenna multiple stream transmission at base stations without direct interference between the APs. In addition, the proposed mechanism synchronizes IoT information by deep learning the frequency of use of secret keys according to the number of IoT information by applying the method of distributing secret keys in dimension according to the number of frequency channels of IoT information in order to make the most of the multiple antenna technology.

• Journal of The Korean Association of Information Education
• /
• v.22 no.4
• /
• pp.457-471
• /
• 2018

This study examined student- and school-level background and ICT factors that affected PISA 2015 Collaborative Problem Solving (CPS) for Korean students (4863 students from 142 high schools). A two-level hierarchical linear model (HLM) was analyzed from the basic model (model 1) with no predictors to the final model (model 5) with all predictors. Results showed that first, gender, socioeconomic/cultural backgrounds, cooperation level positively predicted CPS scores while perceived unfairness of teacher negatively predicted the outcome. Second, the more frequently ICT was used for out-of-school learning purposes, the less frequently ICT was used for entertainment purposes, and the less frequently ICT was used in schools, the higher CPS scores were. Considering ICT autonomy and social interaction variables measured for the first time in PISA 2015, students who were more interested in ICT and more autonomous in using ICT devices achieved higher CPS scores. On the other hand, the more students considered ICT important as social interaction, the less they gained CPS scores. Third, in terms of school-level characteristics, the smaller the students behavior detrimental to learning, the higher the teachers perceived positive working environment, and the fewer the number of computers available per student, the higher CPS scores were. To facilitate computer-based collaborative problem-solving competence, it is important for students to have interest and autonomy in using ICT. In addition, the guidelines of ICT use and SW curriculum need to be established in order to increase the effectiveness of using ICT device in school.

• Journal of Korean Society for Atmospheric Environment
• /
• v.34 no.5
• /
• pp.735-747
• /
• 2018

Rainfall depth is an important meteorological information. Generally, high spatial resolution rainfall data such as road-level rainfall data are more beneficial. However, it is expensive to set up sufficient Automatic Weather Systems to get the road-level rainfall data. In this paper, we propose to use deep learning to recognize rainfall depth from road surveillance videos. To achieve this goal, we collect a new video dataset and propose a procedure to calculate refined rainfall depth from the original meteorological data. We also propose to utilize the differential frame as well as the optical flow image for better recognition of rainfall depth. Under the Temporal Segment Networks framework, the experimental results show that the combination of the video frame and the differential frame is a superior solution for the rainfall depth recognition. The final model is able to achieve high performance in the single-location low sensitivity classification task and reasonable accuracy in the higher sensitivity classification task for both the single-location and the multi-location case.

• The KIPS Transactions:PartA
• /
• v.11A no.4
• /
• pp.243-250
• /
• 2004

Monitoring is used to see if a real-time system provides a service on time. Generally, monitoring for real-time focuses on investigating the current status of a real-time system. To support a stable performance of a real-time system, it should have not only a function to see the current status of real-time process but also a function to predict executions of real-time processes, however. The legacy prediction model has some limitation to apply it to a real-time monitoring. First, it performs a static prediction after a real-time process finished. Second, it needs a statistical pre-analysis before a prediction. Third, transition probability and data about clustering is not based on the current data. We propose the execution prediction model based on learning algorithm to solve these problems and apply it to real-time monitoring. This model gets rid of unnecessary pre-processing and supports a precise prediction based on current data. In addition, this supports multi-level prediction by a trend analysis of past execution data. Most of all, We designed the model to support dynamic prediction which is performed within a real-time process' execution. The results from some experiments show that the judgment accuracy is greater than 80% if the size of a training set is set to over 10, and, in the case of the multi-level prediction, that the prediction difference of the multi-level prediction is minimized if the number of execution is bigger than the size of a training set. The execution prediction model proposed in this model has some limitation that the model used the most simplest learning algorithm and that it didn't consider the multi-regional space model managing CPU, memory and I/O data. The execution prediction model based on a learning algorithm proposed in this paper is used in some areas related to real-time monitoring and control.