• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 1997.10a
• /
• pp.301-308
• /
• 1997

The learning speed of the neural networks, the most important factor in applying to real problems, greatly depends on the learning rate of the networks, Three approaches-empirical, deterministic, and stochastic ones-have been proposed to date. We proposed a new learning-rate selection algorithm using an evolutionary programming search scheme. Even though the performance of our method showed better than those of the other methods, it was found that taking much time for selecting evolutionary learning rates made the performance of our method degrade. This was caused by using static intervals (called static windows) in order to update learning rates. Out algorithm with static windows updated the learning rates showed good performance or didn't update the learning rates even though previously updated learning rates shoved bad performance. This paper introduce a window control scheme to avoid such problems. With the window control scheme, our algorithm try to update the learning ra es only when the learning performance is continuously bad during a specified interval. If previously selected learning rates show good performance, new algorithm will not update the learning rates. This diminish the updating time of learning rates greatly. As a result, our algorithm with the window control scheme show better performance than that with static windows. In this paper, we will describe the previous and new algorithm and experimental results.

• Journal of the Korea Society of Computer and Information
• /
• v.25 no.7
• /
• pp.39-46
• /
• 2020

Recently, for designing virtual characters in the battle game field effectively, some methods are very needed to predicate the win rates of the battle of them efficiently. In this paper, we propose a method to solve this problem by combining simulation and machine learning. Firstly, a simulation is used to analyze the win rates of the battle of virtual characters in the battle game. In addition, we apply a regression model based machine learning scheme to predict win rates of the battle of virtual characters according to their abilities. Our experimental results using suggested method show that it is almost no difference between the win rates of the simulation and the prediction results using the machine learning scheme. And also, we can obtain good performance in the experiment using only simple regression based machine learning model.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.15 no.7
• /
• pp.800-804
• /
• 2005

This paper presents a fuzzy learning rule which is the fuzzified version of LVQ(Learning Vector Quantization). This fuzzy learning rule 3 uses fuzzy learning rates. instead of the traditional learning rates. LVQ uses the same learning rate regardless of correctness of classification. But, the new fuzzy learning rule uses the different learning rates depending on whether classification is correct or not. The new fuzzy learning rule is integrated into the improved IAFC(Integrated Adaptive Fuzzy Clustering) neural network. The improved IAFC neural network is both stable and plastic. The iris data set is used to compare the performance of the supervised IAFC neural network 3 with the performance of backprogation neural network. The results show that the supervised IAFC neural network 3 is better than backpropagation neural network.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.11
• /
• pp.142-150
• /
• 1994

This paper deals with fast competitive learning using classified learning rates. The basic idea of the proposed method is to assign a classified learning rate to each weight vector. The weight vector associated with an output node is updated using its own learning rate. Each learning rate is changed only when its corresponding output node wins the competition, and the learning rates of the losing nodes are not changed. The experimental results obtained with image vector quantization show that the proposed method learns more rapidly and yields better quality that conventional competitive learning.

• The Journal of the Korea Contents Association
• /
• v.4 no.3
• /
• pp.99-105
• /
• 2004

This paper proposes optimal learning rates in the gradient descent training of multilayer perceptrons, which are a separate learning rate for weights associated with each neuron and a separate one for assigning virtual hidden targets associated with each training pattern Effectiveness of the proposed error function was demonstrated for a handwritten digit recognition and an isolated-word recognition tasks and very fast learning convergence was obtained.

• Journal of Information Processing Systems
• /
• v.16 no.5
• /
• pp.1048-1063
• /
• 2020

In this study, we propose a reinforcement learning agent to control the data transmission rates of nodes in carrier sensing multiple access with collision avoidance (CSMA/CA)-based wireless networks. We design a reinforcement learning (RL) agent, based on Q-learning. The agent learns the environment using the timeout events of packets, which are locally available in data sending nodes. The agent selects actions to control the data transmission rates of nodes that adjust the modulation and coding scheme (MCS) levels of the data packets to utilize the available bandwidth in dynamically changing channel conditions effectively. We use the ns3-gym framework to simulate RL and investigate the effects of the parameters of Q-learning on the performance of the RL agent. The simulation results indicate that the proposed RL agent adequately adjusts the MCS levels according to the changes in the network, and achieves a high throughput comparable to those of the existing data transmission rate adaptation schemes such as Minstrel.

• Journal of the Korean Statistical Society
• /
• v.36 no.3
• /
• pp.367-372
• /
• 2007

Classification as a tool to extract information from data plays an important role in science and engineering. Among various classification methodologies, support vector machine has recently seen significant developments. The central problem this paper addresses is the accuracy of support vector machine. In particular, we are interested in the situations where fast rates of convergence to the Bayes risk can be achieved by support vector machine. Through learning examples, we illustrate that support vector machine may yield fast rates if the space spanned by an adopted kernel is sufficiently large.

• Journal of the Chungcheong Mathematical Society
• /
• v.34 no.2
• /
• pp.157-168
• /
• 2021

This study uses the benchmark interest rate of the Federal Open Market Committee (FOMC) to predict gold futures prices. For the predictions, we used the support vector machine (SVM) (a machine-learning model) and the long short-term memory (LSTM) deep-learning model. We found that the LSTM method is more accurate than the SVM method. Moreover, we applied the Boruta algorithm to demonstrate that the FOMC benchmark interest rates correlate with gold futures.

• Korea Trade Review
• /
• v.46 no.1
• /
• pp.169-187
• /
• 2021

This paper analyzes the feasibility of using machine learning and deep learning methods to forecast the income and employment using the strategic industries as well as investment, export, and exchange rates. The decision tree, artificial neural network, support vector machine, and deep learning models were used to forecast the income and employment in Busan. The following were the main findings of the comparison of their predictive abilities. First, the decision tree models predict the income and employment well. The forecasting values for the income and employment appeared somewhat differently according to the depth of decision trees and several conditions of strategic industries as well as investment, export, and exchange rates. Second, since the artificial neural network models show that the coefficients are somewhat low and RMSE are somewhat high, these models are not good forecasting the income and employment. Third, the support vector machine models show the high predictive power with the high coefficients of determination and low RMSE. Fourth, the deep neural network models show the higher predictive power with appropriate epochs and batch sizes. Thus, since the machine learning and deep learning models can predict the employment well, we need to adopt the machine learning and deep learning models to forecast the income and employment.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.4
• /
• pp.175-182
• /
• 1994

In this study, we present unvoiced consonant recognition system using characteristic parameters of the phoneme of the each syllable. For the recognition, the characteristic parameters on the time domain such as ZCR, total energy of the consonant region and half region energy of the consonant region, and those on the frequency domain such as the frequency spectrum of the transition region are used. The objective unvoiced consonants in this study are /ㄱ/,/ㄷ/,/ㅂ/,/ㅈ/,/ㅋ/,/ㅌ/,/ㅍ/ and /ㅊ/. Each characteristic parameter of two regions extracted from these segmented unvoiced consonants are used for each recognition system of the region, independently, And complementing two outputs of each other system, the final output is to be produced. The recognition system is implemented using MLP which has learning ability. The recognition simulation results for 112 unvoiced consonant samples are that average recognition rates are 96.4$\%$ under 80$\%$ learning rates and 93.7$\%$ under 60$\%$ learning rates.