• International Journal of Control, Automation, and Systems
• /
• 제4권6호
• /
• pp.725-735
• /
• 2006

This paper deals with the problem of self-learning cooperative motion control for the pushing task of a humanoid robot in the sagittal plane. A model with 27 linked rigid bodies is developed to simulate the system dynamics. A simple genetic algorithm(SGA) is used to find the cooperative motion, which is to minimize the total energy consumption for the entire humanoid robot body. And the multi-layer neural network based on backpropagation(BP) is also constructed and applied to generalize parameters, which are obtained from the optimization procedure by SGA, in order to control the system.

• E2M - 전기 전자와 첨단 소재
• /
• 제10권8호
• /
• pp.783-793
• /
• 1997

A VLSI implementation of a self-learning neural network integrated circuits using a linearity synapse transistor is investigated. The thickness dependence of oxide current density stress current transient current and channel current has been measured in oxides with thicknesses between 41 and 112 $\AA$, which have the channel width $\times$ length 10 $\times$1${\mu}{\textrm}{m}$, 10 $\times$ 0.3${\mu}{\textrm}{m}$ respectively. The transient current will affect data retention in synapse transistors and the stress current is used to estimate to fundamental limitations on oxide thicknesses. The synapse transistor has represented the neural states and the manipulation which gaves unipolar weights. The weight value of synapse transistor was caused by the bias conditions. Excitatory state and inhitory state according to weighted values affected the drain source current.

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 1996년도 하계학술대회 논문집 B
• /
• pp.1182-1185
• /
• 1996

The objective of this paper is to control a nonlinear dynamical systems based on Self-Organized Distributed Networks (SODN). The learning with the SODN is fast and precise. Such properties are caused from the local learning mechanism Each local network learns only data in a subregion. Methods for indirect adaptive control of nonlinear systems using the SODN is presented. Through extensive simulation, the SODN is shown to be effective for adaptive control of nonlinear dynamic systems.

• 한국전기전자재료학회:학술대회논문집
• /
• 한국전기전자재료학회 2000년도 하계학술대회 논문집
• /
• pp.59-62
• /
• 2000

A VLSI implementation of a self-learning neural network integrated circuits using a linearity synapse transistor is investigated. The thickness dependence of oxide current density, stress current, transient current and channel current has been measured in oxides with thicknesses between 41 and 112 $\AA$, which have the channel width$\times$length 10$\times$1${\mu}{\textrm}{m}$ respectively. The transient current will affect data retention in synapse transistors and the stress current is used to estimate to fundamental limitations on oxide thicknesses. The synapse transistor has represented the neural states and the manipulation which gave unipolar weights. The weight value of synapse transistor was caused by the bias conditions. Excitatory state and inhitory state according to weighted values affected the drain source current.

• 로봇학회논문지
• /
• 제17권4호
• /
• pp.425-430
• /
• 2022

As robots are considered one of the mainstream digital transformations, robots with machine vision becomes a main area of study providing the ability to check what robots watch and make decisions based on it. However, it is difficult to find a small object in the image mainly due to the flaw of the most of visual recognition networks. Because visual recognition networks are mostly convolution neural network which usually consider local features. So, we make a model considering not only local feature, but also global feature. In this paper, we propose a detection method of a small marker on the object using deep learning and an algorithm that considers global features by combining Transformer's self-attention technique with a convolutional neural network. We suggest a self-attention model with new definition of Query, Key and Value for model to learn global feature and simplified equation by getting rid of position vector and classification token which cause the model to be heavy and slow. Finally, we show that our model achieves higher mAP than state of the art model YOLOr.

• International Journal of Computer Science & Network Security
• /
• 제21권2호
• /
• pp.66-76
• /
• 2021

The quality of the E-learning education in Saudi Arabia has been a major concern by many academicians, especially, and people in general as this platform has not been a priority for education. Not until recently, the world has been impacted by the Covid-19 pandemic, which makes every education institution shifted to the online platform to continue the education for the students. Thus, many studies on the perceptions on the online learning have been carried out, and though many are focusing on the perceptions by the education institutions' faculty and administration, there is a lack in the amount of study performed to analyse the students' perceptions of online learning during the pandemic time. The current study is conducted by utilising qualitative methods in order to collect information and investigate the students' perception regarding online learning during the pandemic Covid-19, based on their individual experiences. A number of fifteen (15) students were selected as respondents for the study, in which structured interviews were conducted by using a convenient sampling technique for data collection. Through the discussion, all of the positive and negative perceptions of online learning, as well as the factors contributing to those perceptions were identified. The results of the study found that the positive perceptions were contributed based on the flexibility, cost-effectiveness, availability of the electronic research databases, and well-designed online classroom interfaces. For the negative perceptions from using online learning platforms, the respondents informed that they were contributed by the lecturer's delayed feedback, lack of technical support by lecturers, low in self-esteem and self-motivation, feel isolated, one-way of educational methods, and poorly-designed class materials. Through the findings, the school's administration and lecturers would be able to know the struggles experienced by the students, and eventually come out with better solutions to improve their teaching methods.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제17권3호
• /
• pp.721-739
• /
• 2023

Rating prediction is an important issue in recommender systems, and its accuracy affects the experience of the user and the revenue of the company. Traditional recommender systems use Factorization Machinesfor rating predictions and each feature is selected with the same weight. Thus, there are problems with inaccurate ratings and limited data representation. This study proposes a deep recommendation model based on self-attention Factorization (SAFMR) to solve these problems. This model uses Convolutional Neural Networks to extract features from user and item reviews. The obtained features are fed into self-attention mechanism Factorization Machines, where the self-attention network automatically learns the dependencies of the features and distinguishes the weights of the different features, thereby reducing the prediction error. The model was experimentally evaluated using six classes of dataset. We compared MSE, NDCG and time for several real datasets. The experiment demonstrated that the SAFMR model achieved excellent rating prediction results and recommendation correlations, thereby verifying the effectiveness of the model.

• 디지털산업정보학회논문지
• /
• 제13권4호
• /
• pp.231-245
• /
• 2017

The purpose of this study is to analyze the trend line through scatter diagram analysis on correlation between SNS addiction tendencies and learning attitudes, figure out the knee point influencing learning attitudes negatively in detail, and examine influence among subareas. To address the goal, study questions are formulated as follows. First, this author did screening on the data of variables measured and analyzed descriptive statistics. Second, this researcher produced the trend line by drawing a scatter diagram in order to analyze correlation between SNS addiction tendencies, withdrawal symptoms, excessive communication, and excessive time wasting, and learning attitudes exploratorily. Third, to explore correlation between self-evaluation, learning participation, and developmental attitudes, the subfactors of learning attitudes related to SNS addiction tendencies, this author drew a scatter diagram and analyzed the threshold of positive and negative correlation. To verify the study questions, the SNS addiction tendency scale and learning attitude scale were applied to 301 university students in Chungcheong area. According to the study results, first, their learning attitudes are influenced by SNS addiction tendencies, excessive communication and excessive time wasting, and they are not influenced by withdrawal symptoms that much. Second, excessive communication, a factor of SNS addiction tendencies, and self-evaluation and developmental attitudes, factors of learning attitudes, show positive correlation to some extent and indicate negative correlation after the threshold. However, excessive communication and learning participation are found to show no correlation. Third, according to the results of examining correlation with learning attitudes by dividing them into excessive communication and excessive time wasting groups with the knee point of 1.40, as the symptom of excessive communication is found more, it influences self-evaluation, learning participation, developmental attitudes, and learning attitudes more negatively in general. The result of this study is expected to provide foundational material necessary to develop educational programs to prevent undergraduates' excessive SNS use and SNS addiction which can be used in the scenes of counseling or education.

• 전자공학회논문지B
• /
• 제29B권8호
• /
• pp.1-6
• /
• 1992

This paper proposed a neural computer architecture for the learning of script character pattern recognition categories. Oriented filter with complex cells preprocess about the input script character, abstracts contour from the character. This contour normalized and inputed to the ART. Top-down attentional and matching mechanisms are critical in self-stabilizing of the code learning process. The architecture embodies a parallel search scheme that updates itself adaptively as the learning process unfolds. After learning ART self-stabilizes, recognition time does not grow as a function of code complexity. Vigilance level shows the similarity between learned patterns and new input patterns. This character recognition system is designed to adaptable. The simulation of this system showed satisfied result in the recognition of the hand written characters.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1991년도 한국자동제어학술회의논문집(국제학술편); KOEX, Seoul; 22-24 Oct. 1991
• /
• pp.1570-1575
• /
• 1991

The feasibility of using an adaptive resonance network (ART2) with unsupervised learning capability for too] wear detection in turning operations is investigated. Specifically, acoustic emission (AE) and cutting force signals were measured during machining, the multichannel AR coefficients of the two signals were calculated and then presented to the network to make a decision on tool wear. If the presented features are significantly different from previously learned patterns associated with a fresh tool, the network will recognize the difference and form a new category m worn tool. The experimental results show that tool wear can be effectively detected with or without minimum prior training using the self-organization property of the ART2 network.