• The Transactions of the Korean Institute of Electrical Engineers
• /
• v.39 no.8
• /
• pp.856-867
• /
• 1990

In this paper, the convergence and the learning accuracy of the back-propagation (BP) method in neural network are investigated by 1) analyzing the reason for decelerating the convergence of BP method and examining the rapid deceleration of the convergence when the learning is executed on the part of sigmoid activation function with the very small first derivative and 2) proposing the modified logistic activation function by defining, the convergence factor based on the analysis. Learning on the output patterns of binary as well as analog forms are tested by the proposed method. In binary output patter, the test results show that the convergence is accelerated and the learning accuracy is improved, and the weights and thresholds are converged so that the stability of neural network can be enhanced. In analog output patter, the results show that with extensive initial transient phenomena the learning error is decreased according to the convergence factor, subsequently the learning accuracy is enhanced.

• Proceedings of the KIEE Conference
• /
• 1992.07a
• /
• pp.480-483
• /
• 1992

Although the back-propagation(BP) algorithm is widely used for its simple structure and easy learning method, it has a drawback of slow convergence rate. In this paper, we propose an algorithm to improve this problem by manipulating the slope parameter of the activation function. The steepest descent method is used in learning the slope parameter, as in the case of weight. The simulation shows that the learning rates of the proposed algorithm is faster than the conventional BP algorithm.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.45 no.6
• /
• pp.16-21
• /
• 2008

This paper introduces a new type of Hopfield neural network using newly developed single-electron devices. In the electrical model of the Hopfield neural network, a single-electron synapse, used as a voltage(or current)-variable resistor, and two stages of single-electron inverters, used as a nonlinear activation function, are simulated with a single-electron circuit simulator using Monte-Carlo method to verily their operation.

• Journal of Physiology & Pathology in Korean Medicine
• /
• v.28 no.1
• /
• pp.59-62
• /
• 2014

Glucose uptake plays a pivotal role in maintaining whole body glucose homeostasis in adipocytes and skeletal muscles. In the present study we have shown that Fructus Piperis Longi Extracts (FPLE) can stimulate glucose uptake in OP9 adipocytes. The increasing effects of FPLE on glucose uptake were inhibited by compound C pretreatment, which means that the glucose uptake effects by FPLE were carried out by AMP-activated protein kinase (AMPK) activation. Further studies revealed that FPLE stimulated glucose transport occurs through a mechanism involving extracellular signal-regulated kinase (ERK1/2) activation.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.3
• /
• pp.143-148
• /
• 2022

In this paper, we employ deep neural network (DNN) to discern Android user interaction data distribution from artificial data distribution. We utilize real Android user interaction trace dataset collected from [1] to evaluate our DNN design. In particular, we use sequential model with 4 dense hidden layers and 1 dense output layer in TensorFlow and Keras. We also deploy sigmoid activation function for a dense output layer with 1 neuron and ReLU activation function for each dense hidden layer with 32 neurons. Our evaluation shows that our DNN design fulfills high test accuracy of at least 0.9955 and low test loss of at most 0.0116 in all cases of artificial data distributions.

• 제어로봇시스템학회:학술대회논문집
• /
• 2000.10a
• /
• pp.16-16
• /
• 2000

This paper describes a new structure re create a pseudo Gaussian function network (PGFN). The activation function of hidden layer does not necessarily have to be symmetric with respect to center. To give the flexibility of the network, the deviation of pseudo Gaussian function is changed according to a direction of given input. This property helps that given function can be described effectively with a minimum number of center by PGFN, The distribution of deviation is represented by level set method and also the loaming of deviation is adjusted based on it. To demonstrate the performance of the proposed network, general problem of function estimation is treated here. The representation problem of continuous functions defined over two-dimensional input space is solved.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2004.10a
• /
• pp.1303-1306
• /
• 2004

We developed a symmetrical upper limb motion trainer for chronic hemiparetic subjects. This trainer enabled the practice of a forearm pronatio $n^ination and wrist flexion/extension. In this study, we have used functional magnetic resonance imaging(fMRI) with the developed symmetrical upper limb motion device, to compare brain activation patterns elicited by flexion/extension wrist movements of control and hemiparetic subject group. In control group, contralateral somatosensory cortex(SMC) and bilateral cerebellum were activated by dominant hand movement(Task 1), while bilateral movements by dominant hand(Task 2) activated the SMC in both cerebral hemispheres and ipsilateral cerebellum. However, in hemiparetic subject group, contralateral supplymentary motor area(SMA) was activated by unaffected hand movement(Task 1), while the activation of bilateral movements by unaffected hand(Task 2) showed only SMA in the undamaged hemisphere. This study, demonstrating the ability to accurately measure activation in both sensory and motor cortex, is currently being extended to patients in clinical applications such as the recovery of motor function after stroke.ke.

• Plant Biotechnology Reports
• /
• v.5 no.1
• /
• pp.19-25
• /
• 2011

Tocopherols belong to the plant-derived poly phenolic compounds known for antioxidant functions in plants and animals. Activation of mitogen-activated protein kinases (MAPK) is a common reaction of plant cells in defense-related signal transduction pathways. We report a novel non-antioxidant function of ${\alpha}$-tocopherol in higher plants linking the physiological role of tocopherol with stress signalling pathways. Pre-incubation of a low concentration of $50{\mu}M$ ${\alpha}$-tocopherol negatively interferes with MAPK activation in elicitor-treated tobacco BY2 suspension culture cells and wounded tobacco leaves, whereas pre-incubated BY2 cells with ${\alpha}$-tocopherol phosphate did not show the inhibitory effect on stimuli-induced MAPK activation. The decreased MAPK activity was neither due to a direct inhibitory effect of ${\alpha}$-tocopherol nor due to the induction of an inhibitory or inactivating activity directly affecting MAPK activity. The data support that the target of ${\alpha}$-tocopherol negatively regulates an upstream component of the signaling pathways that leads to stress dependent MAPK activation.

• Journal of Plant Biotechnology
• /
• v.36 no.3
• /
• pp.294-300
• /
• 2009

Activation tagging that uses T-DNA vectors containing multimerized transcriptional enhancers from the cauliflower mosaic virus (CaMV) 35S gene is a powerful tool to determine gene function in plants. This approach has been successfully applied in screening various types of mutations and cloning the corresponding genes. We generated an activation tagged hairy root pool of ginseng (Panax ginseng C.A. Meyer) in an attempt to isolate genes involved in the biosynthetic pathway of ginsenoside (triterpene saponin), which is known as the major active ingredient of the root. Quantitative and qualitative variation of ginsenoside in activation tagged hairy root lines were profiled using LC/MS. Metabolic profiling data enabled selection of a specific hairy root line which accumulated ginsenoside at a higher level than other lines. The relative expression level of several genes of triterpene biosynthetic pathway in the selected hairy root line was determined by real time RT-PCR. Overall results suggest that the activation tagged ginseng hairy root system described in this study would be useful in isolating genes involved in a complex metabolic pathway from genetically intractable plant species by metabolic profiling.

• Journal of the Korea Society for Simulation
• /
• v.13 no.4
• /
• pp.1-16
• /
• 2004

The modelling and simulation of the activation process for the heart system is meaningful to understand special excitatory and conductive system in the heart and to study cardiac functions because the heart activation conducts through this system. This thesis proposes two dimensional cellular automaton(CA) model for the activation process of the myocardium and conducted simulation by means of discrete time and discrete event algorithm. In the model, cells are classified into anatomically similar characteristic parts of the heart and each of cells has a set of cells with preassigned properties. Each cell in this model has state variables to represent the state of the cell and has some state transition rules to change values of state variables executed by state transition function. The state transition rule is simple as follows. First, the myocardium cell at rest stay in passive state. Second, if any one of neighborhood cell in the myocardium cell is active state then the state is change from passive to active state. Third, if cell's state is an active then automatically go to the refractory state after activation phase. Four, if cell's state is refractory then automatically go to the passive state after refractory phase. These state transition is processed repeatedly in all cells through the termination of simulation.