• 전자공학회논문지B
• /
• 제28B권12호
• /
• pp.27-37
• /
• 1991

In this paper, the novel PLD (Preceding Layer Driven) MLP (Multi Layer Perceptron) neural network model and its learning algorithm is described. This learning algorithm is different from the conventional. This integer weights and hard limit function are used for synaptic weight values and activation function, respectively. The entire learning process is performed by layer-by-layer method. the number of layers can be varied with difficulty of training data. Since the synaptic weight values are integers, the synapse circuit can be easily implemented with CMOS. PLD MLP neural network was applied to English Characters, arbitrary waveform generation and spiral problem.

• 로봇학회논문지
• /
• 제18권1호
• /
• pp.122-126
• /
• 2023

An unexpected occlusion in a real life, not in a laboratory, can be more fatal to neural networks than expected. In addition, it is virtually impossible to create a network that learns all the environmental changes as well as occlusions. Therefore, we propose an alternative approach in which the architecture and number of parameters remain unchanged while adapting to occlusion circumstances. Learning method with the term Conversion Learning classifies them more robustly by converting the weights from various occlusion situations. The experiments on MNIST dataset showed a 3.07 [%p] performance improvement over the baseline CNN model in a situation where most objects are occluded and unknowing what occlusion will appear in advance. The experimental results suggest that Conversion Learning is an efficient method to respond to environmental changes such as occluded images.

• 한국컴퓨터정보학회논문지
• /
• 제23권7호
• /
• pp.81-90
• /
• 2018

In this paper, reducing lense Vignetting effect and adaptive learning rate method are proposed to complement Scribner's neural network for nuc algorithm which is the effective algorithm in statistic SBNUC algorithm. Proposed reducing vignetting effect method is updated weight and bias each differently using different cost function. Proposed adaptive learning rate for updating weight and bias is using sobel edge detection method, which has good result for boundary condition of image. The ordinary statistic SBNUC algorithm has problem to compensate lense vignetting effect, because statistic algorithm is updated weight and bias by using gradient descent method, so it should not be effective for global weight problem same like, lense vignetting effect. We employ the proposed methods to Scribner's neural network method(NNM) and Torres's reducing ghosting correction for neural network nuc algorithm(improved NNM), and apply it to real-infrared detector image stream. The result of proposed algorithm shows that it has 10dB higher PSNR and 1.5 times faster convergence speed then the improved NNM Algorithm.

• 대한전기학회논문지
• /
• 제40권2호
• /
• pp.224-233
• /
• 1991

The multilayer expanson of single layer NN (Neural Network) was needed to solve the linear seperability problem as shown by the classic example using the XOR function. The EBP (Error Back Propagation ) learning rule is often used in multilayer Neural Networks, but it is not without its faults: 1)D.Rimmelhart expanded the Delta Rule but there is a problem in obtaining Ca from the linear combination of the Weight matrix N between the hidden layer and the output layer and H, wich is the result of another linear combination between the input pattern and the Weight matrix M between the input layer and the hidden layer. 2) Even if using the difference between Ca and Da to adjust the values of the Weight matrix N between the hidden layer and the output layer may be valid is correct, but using the same value to adjust the Weight matrixd M between the input layer and the hidden layer is wrong. Recognitron III was proposed to solve these faults. According to simulation results, since Recognitron III does not learn the three layer NN itself, but divides it into several single layer NNs and learns these with learning patterns, the learning time is 32.5 to 72.2 time faster than EBP NN one. The number of patterns learned in a EBP NN with n input and output cells and n+1 hidden cells are 2**n, but n in Recognitron III of the same size. [5] In the case of pattern generalization, however, EBP NN is less than Recognitron III.

• 한국컴퓨터정보학회논문지
• /
• 제26권10호
• /
• pp.53-59
• /
• 2021

본 논문에서는 어떠한 장치를 착용하지 않고 덤벨의 무게감을 인지하여 덤벨 운동을 할 수 있는 시스템을 제안한다. 가상현실 기술이 발전하면서 가상세계에서 현실 세계의 물리적 피드백을 시뮬레이션 하는 연구가 많이 진행하고 있다. 노인들에게 있어 재활 운동은 정확한 모션 인식이 중요하다. 또한, 노인들은 무거운 덤벨을 들 수가 없다. 본 논문에서는 재활 운동을 위해 실시간으로 입력되는 동작 데이터를 머신러닝을 활용하여 정확성을 판별하였다. 실험으로서 한손 덤벨 운동, 양손 덤벨 운동, 어깨 덤벨 운동 3가지를 실험하였으며, 재활 운동이 정확히 이루어지는가에 대해 실험하였다. 가장 안정적인 운동은 양손을 활용한 덤벨 운동이었으며 가장 적은 오차율을 나타내었다. 마지막으로, 가상에서 실제 적용하기 위해 가상 체육관 콘텐츠를 개발하였다.

• 대한한의학회지
• /
• 제41권2호
• /
• pp.58-79
• /
• 2020

Objectives: The purpose of this study is to predict the weight loss by applying machine learning using real-world clinical data from overweight and obese adults on weight loss program in 4 Korean Medicine obesity clinics. Methods: From January, 2017 to May, 2019, we collected data from overweight and obese adults (BMI≥23 kg/m2) who registered for a 3-month Gamitaeeumjowi-tang prescription program. Predictive analysis was conducted at the time of three prescriptions, and the expected reduced rate and reduced weight at the next order of prescription were predicted as binary classification (classification benchmark: highest quartile, median, lowest quartile). For the median, further analysis was conducted after using the variable selection method. The data set for each analysis was 25,988 in the first, 6,304 in the second, and 833 in the third. 5-fold cross validation was used to prevent overfitting. Results: Prediction accuracy was increased from 1^st to 2^nd and 3^rd analysis. After selecting the variables based on the median, artificial neural network showed the highest accuracy in 1^st (54.69%), 2^nd (73.52%), and 3^rd (81.88%) prediction analysis based on reduced rate. The prediction performance was additionally confirmed through AUC, Random Forest showed the highest in 1^st (0.640), 2^nd (0.816), and 3^rd (0.939) prediction analysis based on reduced weight. Conclusions: The prediction of weight loss by applying machine learning showed that the accuracy was improved by using the initial weight loss information. There is a possibility that it can be used to screen patients who need intensive intervention when expected weight loss is low.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 한국자동제어학술회의논문집(국내학술편); 포항공과대학교, 포항; 24-26 Oct. 1996
• /
• pp.936-939
• /
• 1996

In this paper, two specially designed associative mapping memories, called Associative Mapping Elements(AME) and Multiple-Digit Overlapping AME(MDO-AME), are presented for learning of nonlinear functions including kinematics and dynamics of robot manipulators. The proposed associative mapping memories consist of associative mapping rules(AMR) and weight update rules(WUR) which guarantee generalization and specialization of input-output relationship of learned nonlinear functions. Two simulation results, one for supervised learning and the other for unsupervised learning, are given to demonstrate the effectiveness of the proposed associative mapping memories.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1990년도 한국자동제어학술회의논문집(국내학술편); KOEX, Seoul; 26-27 Oct. 1990
• /
• pp.36-40
• /
• 1990

Conventional back propagation learning is generally characterized by slow and rather inaccurate learning which makes it difficult to use in control applications. A new multilayer perception architecture and its learning algorithm is proposed that consists of a Kohonen front layer followed by a back propagation network. The Kohonen layer selects a subset of the hidden layer neurons for local tuning. This architecture has been tested on the inverse kinematic solution of robot manipulator while demonstrating its fast and accurate learning capabilities.

• 정보학연구
• /
• 제4권1호
• /
• pp.71-79
• /
• 2001

본 논문에서는 역 전파 신경회로망의 학습파라미터를 발생한 오차에 따라서 유동성 있게 갱신할 수 있고 이 학습알고리즘의 효율을 향상시킬 수 있는 초기연결강도 설정 방법을 제안하였다. 제안한 알고리즘은 국소 점을 벗어날 수 있는 것으로 기대되고, 수렴환경에 알맞은 초기 연결강도 발생을 설정할 수 있다. 모의실험에서는 세 가지의 학습패턴을 가지고 실험하였다. 첫 번째 3-패리티 문제에 대한 학습을 수행하였고, 두 번째는 $7{\times}5$ 알파벳 영문자 폰트에 대한 학습이고 세 번째는 필기체 숫자 및 한글의 기본 획에 적용하였다. 모의실험결과, 제안된 방법은 기존의 표준 역 전파 알고리즘에 비하여 약 27%~57.2%정도 학습효율이 향상됨을 고찰하였다

• International Journal of Contents
• /
• 제13권3호
• /
• pp.25-31
• /
• 2017

With tremendous advancement of information and communication technologies, mobile learning systems have been widely adopted in language learning contexts, and several frameworks have been developed for identifying and categorizing different factors of mobile-assisted language learning (MALL). However, pre-existing frameworks have limitations when evaluating the importance level of criteria. The purpose of this study is to develop a comprehensive hierarchical framework for identifying and categorizing success factors of MALL and prioritizing them according to the importance level. To do that, AHP method is used to quantitatively estimate weight values of MALL criteria. Results reveal that the priority of MALL criteria is ordered as follows: content, system, learner, language learning. Local weights of each criterion are also analyzed; for example, usefulness, accuracy, and authenticity are critical factors for improving MALL contents. Ease of use and mobility of MALL systems are also considered more critical than other systematic factors. In addition, availability of immediate feedback and self-directness has the highest weight values of importance. The findings of the study are discussed regarding hierarchical orders of MALL criteria and conclude that successful MALL implementation may be achieved if related elements are diversely measured and evaluated. Pedagogical implications and suggestions for further research are also presented.