• Journal of KIISE:Software and Applications
• /
• v.31 no.7
• /
• pp.859-868
• /
• 2004

We propose a method of selecting initial training examples for active learning so that it can reach high accuracy faster with fewer further queries. Our method is based on the assumption that an active learner can reach higher performance when given an initial training set consisting of diverse and typical examples rather than similar and special ones. To obtain a good initial training set, we first cluster examples by using k-means clustering algorithm to find groups of similar examples. Then, a representative example, which is the closest example to the cluster's centroid, is selected from each cluster. After these representative examples are labeled by querying to the user for their categories, they can be used as initial training examples. We also suggest a method of using the centroids as initial training examples by labeling them with categories of corresponding representative examples. Experiments with various text data sets have shown that the active learner starting from the initial training set selected by our method reaches higher accuracy faster than that starting from randomly generated initial training set.

• Journal of the Korean Society for Aviation and Aeronautics
• /
• v.27 no.1
• /
• pp.34-42
• /
• 2019

Ten AOC(Air Operation Certificate) holders certified by MOLIT of Korea are operating their own maintenance training program. Though the maintenance training program is approved by the same authority, the contents of the program are different even in the mandatory training courses among AOC holders. The survey interview showed that the maintenance training in mandatory training should have the same contents and requirements. Throughout the survey and focus group discussion, this paper suggests the list and contents of the initial mandatory training and the list, contents and interval for the recurrent mandatory training. This paper also suggests how to implement the on-line training program for recurrent mandatory training to keep the quality of the airline maintenance training program.

• Journal of Advanced Navigation Technology
• /
• v.27 no.3
• /
• pp.269-273
• /
• 2023

Based on the K-UAM roadmap, this study was conducted to select major education and training subjects necessary for the composition of the initial UAM pilot education and training curriculum. Currently, UAM aircraft have similar characteristics to rotorcraft that can take off and land vertically around VTOL functions. Therefore, in this study, the Rotary Wing Pilot Training Curriculum of the Army Aviation School, which represents Rotary Wing Flight Education, was selected as a comparative group, and education and training subjects for initial UAM pilot training were selected. First, a hierarchical structure for the AHP survey was designed based on the Army's rotorcraft pilot education and training subjects, and the AHP survey was conducted by selecting experts from each class. If the education and training subjects given as priorities through AHP analysis are applied to initial UAM pilot training, it is expected to contribute to the effect of education and training and ultimately to the safe operation of UAM.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.4
• /
• pp.145-154
• /
• 1994

This paper proposes an efficient method for improving the training performance of the neural network by using a hybrid of a stochastic approximation and a backpropagation algorithm. The proposed method improves the performance of the training by appliying a global optimization method which is a hybrid of a stochastic approximation and a backpropagation algorithm. The approximate initial point for a stochastic approximation and a backpropagation algorihtm. The approximate initial point for fast global optimization is estimated first by applying the stochastic approximation, and then the backpropagation algorithm, which is the fast gradient descent method, is applied for a high speed global optimization. And further speed-up of training is made possible by adjusting the training parameters of each of the output and the hidden layer adaptively to the standard deviation of the neuron output of each layer. The proposed method has been applied to the parity checking and the pattern classification, and the simulation results show that the performance of the proposed method is superior to that of the backpropagation, the Baba's MROM, and the Sun's method with randomized initial point settings. The results of adaptive adjusting of the training parameters show that the proposed method further improves the convergence speed about 20% in training.

• Phonetics and Speech Sciences
• /
• v.5 no.3
• /
• pp.31-38
• /
• 2013

This study investigated the influence of English vowel pronunciation training to English word-initial stop pronunciation. For that purpose, VOT values of English stops produced by twenty Korean English learners(five Youngnam dialect male speakers, five Youngnam dialect female speakers, five Kangwon dialect male speakers, and five Kangwon dialect female speakers) were measured using the Speech Analyzer and their post-training production was compared with their pre-training production. The result shows that post-training VOT values of voiced stops became closer to those of native English speakers in all four groups. Hence, it can be inferred that vowel pronunciation training is effective for correcting pronunciation of voiced vowels by analyzing the change of the quality of following vowels(especially low vowels) and the degree of giving stress.

• Journal of Intelligence and Information Systems
• /
• v.5 no.1
• /
• pp.95-101
• /
• 1999

In this paper, a comparison between different methods to combine predictions from neural networks will be given. These methods are bagging, bumping, and balancing. Those are based on the analysis of the ensemble generalization error into an ambiguity term and a term incorporating generalization performances of individual networks. Neural Networks and AI machine learning models are prone to overfitting. A strategy to prevent a neural network from overfitting, is to stop training in early stage of the learning process. The complete data set is spilt up into a training set and a validation set. Training is stopped when the error on the validation set starts increasing. The stability of the networks is highly dependent on the division in training and validation set, and also on the random initial weights and the chosen minimization procedure. This causes early stopped networks to be rather unstable: a small change in the data or different initial conditions can produce large changes in the prediction. Therefore, it is advisable to apply the same procedure several times starting from different initial weights. This technique is often referred to as training ensembles of neural networks. In this paper, we presented a comparison of three statistical methods to prevent overfitting of neural network.

• Journal of the Korean Society of Physical Medicine
• /
• v.10 no.4
• /
• pp.49-57
• /
• 2015

PURPOSE: This study aimed to investigate the effect of emphasized initial contact gait training on balance and gait ability in hemiplegia patients. METHODS: Twenty-four hemiplegic patients were randomly allocated to an experimental group or control group. All participants received 30-min neurodevelopmental treatment. Furthermore, the experimental group received initial contact-emphasized auditory feedback gait training, whereas the control group received gait training without auditory feedback. The intervention was performed 3 times per week, 20 min per each time, for a total of 6 weeks. Balance was assessed using the center of pressure path length, center of pressure velocity, and limitation of stability path length, whereas gait ability was assessed using the 10-m walking test and functional gait assessment. RESULTS: In both groups, center of pressure path length and center of pressure velocity significantly decreased after training. Compared to the control group, the experimental group showed a 10% significant improvement (p<.05). In the limitation of stability path length of both sides, the experimental group showed a significant increase compared to that before intervention. Compared to the control group, the experimental group showed a 7% significant improvement in results of the 10-m walking test and functional gait assessment (p<.05). CONCLUSION: Emphasized Initial contact gait training is considered an effective treatment for improving gait ability and balance ability in stroke patients.

• Proceedings of the IEEK Conference
• /
• 2002.07c
• /
• pp.1901-1904
• /
• 2002

A good performance on communication systems is obtained by decreasing the length of training sequence In the initial stage of adaptive equalization. This paper presents a new approach to accomplish this, with the use of a training adaptive equalizer. The approach is based on combining the training and tracking modes, in which the training equalizer is updated by the LMS algorithm with the training sequence and then updated by a blind algorithm. By computer simulations, it is shown that a class of the proposed equalizers provides better performance than the conventional training equalizer.

• Proceedings of the IEEK Conference
• /
• 2001.06c
• /
• pp.173-176
• /
• 2001

This paper proposes training algorithms of neuro-fuzzy systems. First, we introduce a structure training algorithm, which produces the necessary number of hidden nodes from training data. From this algorithm, initial fuzzy rules are also obtained. Second, the parameter training algorithm using evolution strategy is introduced. In order to show their usefulness, we apply our neuro-fuzzy system to a nonlinear system identification problem. It was found from experiments that proposed training algorithms works well.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.54 no.7
• /
• pp.323-327
• /
• 2005

This paper presents an operator training simulator for power system restoration against massive black-out. The system is designed especially focused on the generality and convenient setting up for initial condition of simulation. The former is accomplished by using power flow calculation methodology, and PSS/E data is used to define the initial situation. The proposed simulator consists of three major components - the power flow(PF) module, data conversion(COW) module and GU subsystem. PF module calculates power flow, and then checks overvoltage of buses and overflow of lines. COW module composes an Y-Bus array and a data base at each restoration action. The initial Y-Bus array is constructed from PSS/E data. The user friendly GUI subsystem is developed including graphic editor and built-in operation manual. As a result, the maximum processing time for one step operation is 15 seconds, which is adequate for training purpose. Comparison with PSS/E simulation proves the accuracy and reliability of the training system.