• Communications for Statistical Applications and Methods
• /
• v.28 no.3
• /
• pp.281-294
• /
• 2021

In this paper, we propose an adaptive regression method based on the single-layer neural network structure. We adopt a symmetric activation function as units of the structure. The activation function has a flexibility of its form with a parametrization and has a localizing property that is useful to improve the quality of estimation. In order to provide a spatially adaptive estimator, we regularize coefficients of the activation functions via ℓ₁-penalization, through which the activation functions to be regarded as unnecessary are removed. In implementation, an efficient coordinate descent algorithm is applied for the proposed estimator. To obtain the stable results of estimation, we present an initialization scheme suited for our structure. Model selection procedure based on the Akaike information criterion is described. The simulation results show that the proposed estimator performs favorably in relation to existing methods and recovers the local structure of the underlying function based on the sample.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.15 no.3
• /
• pp.1724-1733
• /
• 2014

This paper presents a pattern recognition analysis of two spirals problem and optimization of Cascade Correlation learning algorithm using in combination with a non-monotone function as CosExp(cosine-modulated symmetric exponential function) and a monotone function as sigmoid function. In addition, the algorithm's optimization is attempted. By using genetic algorithms the optimization of the algorithm will attempt. In the first experiment, by using CosExp activation function for candidate neurons of the learning algorithm is analyzed the recognized pattern in input space of the two spirals problem. In the second experiment, CosExp function for output neurons is used. In the third experiment, the sigmoid activation functions with various parameters for candidate neurons in 8 pools and CosExp function for output neurons are used. In the fourth experiment, the parameters are composed of 8 pools and displacement of the sigmoid function to determine the value of the three parameters is obtained using genetic algorithms. The parameter values applied to the sigmoid activation functions for candidate neurons are used. To evaluate the performance of these algorithms, each step of the training input pattern classification shows the shape of the two spirals. In the optimizing process, the number of hidden neurons was reduced from 28 to15, and finally the learning algorithm with 12 hidden neurons was optimized.

• 제어로봇시스템학회:학술대회논문집
• /
• 2001.10a
• /
• pp.43.3-43
• /
• 2001

In conventional RBF network, the activation functions of hidden layers generally are symmetric functions like gaussian function. This has been considered to be one of the limiting factors for the network to speed up learning of actuately describing a given function. To avoid this criticism, we propose a pseudo gaussian function (PGF) whose deviation is changed according to the direction of incoming pattern. This property helps to estimate the given function more effectively with a minimal number of centers because of its flexibility of functional representation. A level set method is used to describe the asymmetric shape of deviation of the pseudo gaussian function. To demonstrate the performance of the proposed network ...

• 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.

• Physical Therapy Korea
• /
• v.27 no.3
• /
• pp.191-198
• /
• 2020

Background: Stroke patients experience multiple dysfunctions that include motor and sensory impairments. Therefore, new intervention methods require a gradational approach depending on functional levels of a stroke patient's activity and should include cognition treatment to allow for a patient's active participation in rehabilitation. Objects: This study investigates the effect of integrated revision of electrical sensory stimulation, which stimulates somatosensory and action observation training, which is synchronized cognition intervention method on stroke patients' functions. Methods: Twenty-one stroke patients were randomized into two groups. The two groups underwent twenty minutes of intervention five times a week for three weeks. This study used an electromyogram to evaluate symmetric muscle activation of lower extremities and muscle onset time when performing sit to stand before and after intervention. A weight-bearing ratio was used to evaluate the weight-bearing of the affected side in a sit to standing. To evaluate sit to stand performance ability, this study performed five timed sit to stand tests. Results: The two groups both showed statistically significant improvement in muscle onset time of lower extremity, static balance ability in a standing position, and sit to stand performance after the intervention (p < 0.05). In addition, the action observation and synchronized electrical sensory stimulation group showed significant improvement in symmetric muscle activation of lower extremities and weight-bearing ratio of the affected side (p < 0.05). Conclusion: action observation and synchronized electrical sensory stimulation (AOT with ESS) can have positive effects on a stroke patient's sit to stand performance, and the intervention method that provides integrated AOT with ESS can be used as new nervous system intervention program.

• Progress in Medical Physics
• /
• v.16 no.1
• /
• pp.1-9
• /
• 2005

The effect of the developed symmetric upper extremity motion trainer on the cortical activation pattern was investigated in three chronic hemiparetic patients using both fMRI and Fugl-Meyer test. The training program was performed at 1 hr/day, 5 days/week during 6 weeks. Fugl-Meyer tests were performed every two weeks during the training. fMRI was performed at 3T scanner with wrist flexion-extension in two different tasks before and after the training program: the only unaffected hand movement (Task 1) and passive movements of affected hand by the active movement of unaffected hand (Task 2). fMRI studies in Task 1 showed that cortical activations decreased in ipsilateral SMC but increased in contralateral SMC. Task 2 showed cortical reorganizations in bilateral SMC, PMA and SMA. Therefore, it seems that the cortical reorganization in chronic hemiparetic patients can be induced by the training with the developed symmetric upper extremity motion trainer.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.1
• /
• pp.1-9
• /
• 2019

In this paper, we propose a method of generating a 2D gray image from LiDAR 3D reflection intensity. The proposed method uses the Fully Convolutional Network (FCN) to generate the gray image from 2D reflection intensity which is projected from LiDAR 3D intensity. Both encoder and decoder of FCN are configured with several convolution blocks in the symmetric fashion. Each convolution block consists of a convolution layer with $3{\times}3$ filter, batch normalization layer and activation function. The performance of the proposed method architecture is empirically evaluated by varying depths of convolution blocks. The well-known KITTI data set for various scenarios is used for training and performance evaluation. The simulation results show that the proposed method produces the improvements of 8.56 dB in peak signal-to-noise ratio and 0.33 in structural similarity index measure compared with conventional interpolation methods such as inverse distance weighted and nearest neighbor. The proposed method can be possibly used as an assistance tool in the night-time driving system for autonomous vehicles.

• Journal of Korea Society of Industrial Information Systems
• /
• v.28 no.5
• /
• pp.77-87
• /
• 2023

Wind turbines playing a critical role in renewable energy generation, accurately assessing their operational status is crucial for maximizing energy production and minimizing downtime. This study conducts a comparative analysis of different neural network models for wind turbine condition diagnosis, evaluating their effectiveness using a dataset containing sensor measurements and historical turbine data. The study utilized supervisory control and data acquisition data, collected from 2 MW doubly-fed induction generator-based wind turbine system (Model HQ2000), for the analysis. Various neural network models such as artificial neural network, long short-term memory, and recurrent neural network were built, considering factors like activation function and hidden layers. Symmetric mean absolute percentage error were used to evaluate the performance of the models. Based on the evaluation, conclusions were drawn regarding the relative effectiveness of the neural network models for wind turbine condition diagnosis. The research results guide model selection for wind turbine condition diagnosis, contributing to improved reliability and efficiency through advanced neural network-based techniques and identifying future research directions for further advancements.