• Journal of Institute of Control, Robotics and Systems
• /
• v.13 no.4
• /
• pp.309-314
• /
• 2007

This paper presents an adaptive neural computation algorithm for multi-layered neural networks which are applied to identify the characteristic function of dynamic systems. The main feature of the proposed algorithm is that the initial learning rate for the employed neural network is assigned systematically, and also the assigned learning rate can be adjusted empirically for effective neural leaning. By employing the approach, enhanced modeling of dynamic systems is possible. The effectiveness of this approach is veri tied by simulations.

• Proceedings of the KIEE Conference
• /
• 1996.07b
• /
• pp.1371-1373
• /
• 1996

A clustering algorithms with dynamic adjustment of learning rate for GBFCM(Gradient Based FCM) is proposed in this paper. This algorithm combines two idea of dynamic K-means algorithms and GBFCM : learning rate variation with entropy concept and continuous membership grade. To evaluate dynamic GBFCM, we made comparisons with Kohonen's Self-Organizing Map over several tutorial examples and image compression. The results show that DGBFCM(Dynamic GBFCM) gives superior performance over Kohonen's algorithm in terms of signal-to-noise.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2021.10a
• /
• pp.555-556
• /
• 2021

We propose a novel method based on a deep learning model with an optimized dynamic decay learning rate and improved model structure to achieve fast and accurate classification of fashion clothing images.

• Journal of Sensor Science and Technology
• /
• v.30 no.2
• /
• pp.76-81
• /
• 2021

In this study, we propose a machine vision system with a high object recognition rate. By utilizing a multiple-exposure image sensing technique, the proposed deep learning-based machine vision system can cover a wide light intensity range without further learning processes on the various light intensity range. If the proposed machine vision system fails to recognize object features, the system operates in a multiple-exposure sensing mode and detects the target object that is blocked in the near dark or bright region. Furthermore, short- and long-exposure images from the multiple-exposure sensing mode are synthesized to obtain accurate object feature information. That results in the generation of a wide dynamic range of image information. Even with the object recognition resources for the deep learning process with a light intensity range of only 23 dB, the prototype machine vision system with the multiple-exposure imaging method demonstrated an object recognition performance with a light intensity range of up to 96 dB.

• Journal of Korea Multimedia Society
• /
• v.6 no.5
• /
• pp.788-796
• /
• 2003

This paper proposes using dynamic momentum for squential learning method. Using The dynamic momentum improves convergence speed and performance by the variable momentum, also can identify it in the RMSE(root mean squared error). The proposed method is reflected using variable momentum according to current state. While static momentum is equally influenced on the whole, dynamic momentum algorithm can control the convergence rate and performance. According to the variable change of momentum by training. Unlike former classification and regression problems, this paper confirms both performance and regression rate of the dynamic momentum. Using RMSE(root mean square error ), which is one of the regression methods. The proposed dynamic momentum has been applied to the kernel adatron and kernel relaxation as the new sequential learning method of support vector machine presented recently. In order to show the efficiency of the proposed algorithm, SONAR data, the neural network classifier standard evaluation data, are used. The simulation result using the dynamic momentum has a better convergence rate, performance and RMSE than those using the static moment, respectively.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.49 no.1
• /
• pp.41-47
• /
• 2000

A better tracking performance can be achieved, if visual sensors such as CCD cameras are used in controling a robot manipulator, than when only relative sensors such as encoders are used. However, for precise visual servoing of a robot manipulator, an expensive vision system which has fast sampling rate must be used. Moreover, even if a fast vision system is implemented for visual servoing, one cannot get a reliable performance without use of robust and stable inner joint servo-loop. In this paper, we propose a dynamic control scheme for robot manipulators with eye-in-hand camera configuration, where a dynamic learning controller is designed to improve the tracking performance of robotic system. The proposed control scheme is implemented for tasks of tracking moving objects and shown to be robust to parameter uncertainty, disturbances, low sampling rate, etc.

• Journal of Korea Multimedia Society
• /
• v.3 no.4
• /
• pp.416-423
• /
• 2000

The existing self-organizing feature map of Kohonen has weakpoint that need too much input patterns in order to converse into the learning rate and equilibrium state when it trains. Making up for the current weak point, B.Bavarian suggested the method of that distributed the learning rate such as Gaussian function. However, this method has also a disadvantage which can not achieve the right self-organizing. In this paper, we proposed the method of improving the convergence speed and the convergence rate of self-organizing feature map converting the Gaussian function into dynamic approximate curve used in when trains the self-organizing feature map.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.23 no.3
• /
• pp.145-151
• /
• 2023

3D-CNN is one of the deep learning techniques for learning time series data. Such three-dimensional learning can generate many parameters, so that high-performance machine learning is required or can have a large impact on the learning rate. When learning dynamic hand-gestures in spatiotemporal domain, it is necessary for the improvement of the efficiency of dynamic hand-gesture learning with 3D-CNN to find the optimal conditions of input video data by analyzing the learning accuracy according to the spatiotemporal change of input video data without structural change of the 3D-CNN model. First, the time ratio between dynamic hand-gesture actions is adjusted by setting the learning interval of image frames in the dynamic hand-gesture video data. Second, through 2D cross-correlation analysis between classes, similarity between image frames of input video data is measured and normalized to obtain an average value between frames and analyze learning accuracy. Based on this analysis, this work proposed two methods to effectively select input video data for 3D-CNN deep learning of dynamic hand-gestures. Experimental results showed that the learning interval of image data frames and the similarity of image frames between classes can affect the accuracy of the learning model.

• Journal of the Korea Society of Computer and Information
• /
• v.19 no.9
• /
• pp.33-38
• /
• 2014

The FCM based hybrid RBF network is a heterogeneous learning network model that applies FCM algorithm between input and middle layer and applies Max_Min algorithm between middle layer and output. The Max-Min neural network uses winner nodes of the middle layer as input but shows inefficient learning in performance when the input vector consists of too many patterns. To overcome this problem, we propose a dynamic learning rate control based on fuzzy logic. The proposed method first classifies accurate/inaccurate class with respect to the difference between target value and output value with threshold and then fuzzy membership function and fuzzy decision logic is designed to control the learning rate dynamically. We apply this proposed RBF network to the character recognition problem and the efficacy of the proposed method is verified in the experiment.

• The Transactions of the Korean Institute of Electrical Engineers A
• /
• v.48 no.10
• /
• pp.1231-1238
• /
• 1999

This paper presents an approach for the optimal heating load Identification using Diagonal Recurrent Neural Networks(DRNN). In this paper, the DRNN captures the dynamic nature of a system and since it is not fully connected, training is much faster than a fully connected recurrent neural network. The architecture of DRNN is a modified model of the fully connected recurrent neural network with one hidden layer. The hidden layer is comprised of self-recurrent neurons, each feeding its output only into itself. In this study, A dynamic backpropagation (DBP) with delta-bar-delta learning method is used to train an optimal heating load identifier. Delta-bar-delta learning method is an empirical method to adapt the learning rate gradually during the training period in order to improve accuracy in a short time. The simulation results based on experimental data show that the proposed model is superior to the other methods in most cases, in regard of not only learning speed but also identification accuracy.