• The Journal of the Acoustical Society of Korea
• /
• v.39 no.6
• /
• pp.505-514
• /
• 2020

Since a large amount of training data is typically needed to train Deep Neural Networks (DNNs), a parallel training approach is required to train the DNNs. The Stochastic Gradient Descent (SGD) algorithm is one of the most widely used methods to train the DNNs. However, since the SGD is an inherently sequential process, it requires some sort of approximation schemes to parallelize the SGD algorithm. In this paper, we review various efforts on parallelizing the SGD algorithm, and analyze the computational overhead, communication overhead, and the effects of the approximations.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.2 no.2
• /
• pp.159-165
• /
• 2002

This paper proposes a GA and Gradient Descent Method-based method for choosing an appropriate set of fuzzy rules for classification problems. The aim of the proposed method is to fond a minimum set of fuzzy rules that can correctly classify all training patterns. The number of inference rules and the shapes of the membership functions in the antecedent part of the fuzzy rules are determined by the genetic algorithms. The real numbers in the consequent parts of the fuzzy rules are obtained through the use of the descent method. A fitness function is used to maximize the number of correctly classified patterns, and to minimize the number of fuzzy rules. A solution obtained by the genetic algorithm is a set of fuzzy rules, and its fitness is determined by the two objectives, in a combinatorial optimization problem. In order to demonstrate the effectiveness of the proposed method, computer simulation results are shown.

• Korean Journal of Cognitive Science
• /
• v.6 no.3
• /
• pp.25-40
• /
• 1995

This paper proposes a generating method of fuzzy rules by genetic and descent method (GAGDM),and its applied to classification problems.The number of inference rules and the shapes of membership function in the antecedent part are detemined by applying the genetic algorithm,and the real numbers of the consequent parts are derived by using the descent method.The aim of the proposed method is to generation a minmun set of fuzzy rules that can correctly classify all training patterns,and fiteness function of GA defined by the aim of th proposed method.Finally,in order to demonstrate the effectiveness of the present method,simulation results are shown.

• Korean Journal of Acupuncture
• /
• v.28 no.4
• /
• pp.177-198
• /
• 2011

Objectives : The purpose of this study is to examine the theory of mind and body practice in Dan-Jeon-Ju-Seon from the point of views of Taoism, Buddhism, and Oriental medicine. Methods : An ideological background and development of Dan-Jeon-Ju-Seon was first examined. Then, the definition of, other descriptions of, and various locations of, Dan-Jeon were investigated. In addition, the theory of Qi movement of Shui Sheng Huo Jiang (ascent of water Chi and descent of fire Chi) in Dan-Jeon-Ju-Seon was taken into consideration from perspectives on the thought of Taoist Nei Dan (internal alchemy) and Oriental medicine. Finally, the characteristics of mind and body practice in Dan-Jeon-Ju-Seon. Results & Conclusions : Dan-Jeon in Dan-Jeon-Ju-Seon consists of three parts, upper, middle, and lower Dan-Jeon, which is related to Jing (sperm, essence) Qi (breath, eneregy) Shen (spirit, intellect) of our body. Jing Qi Shen is a crucial part in our mind and body, mind and body are connected by energy, and the energy flow is possible by ascent of water Chi and descent of fire Chi. Ultimately, Dan-Jeon-Ju-Seon is a method of practice to keep one's mind and body healthy, and its purpose is to do timeless meditation in our daily lives regardless of time and place.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.15 no.2
• /
• pp.127-133
• /
• 2004

A feedforward amplifier, which is composed of several components, is an open loop system. Therefore, feedforward amplifiers are apt to deteriorate its performance according to the environmental changes even though the cancellation performance and the linearization bandwidth of feedforward systems are superior to other linearization methods. A control method is needed for maintaining the original performance of feedforward amplifiers or to keep the desired performance within a little error bounds. In this paper, an adaptive control method using the steepest descent algorithm, which has a good convergence characteristic and is easy to implement, is suggested. The characteristics of the suggested control method compare with the characteristics of other control methods and the simulation results are presented.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.10 no.2
• /
• pp.276-282
• /
• 2006

PID controller has been widely used in industry. Because it has a simple structure and robustness to modeling error. But it is difficult to have uniformly good control performance in system parameters variation or different velocity command. In this paper, we propose an adaptive PID controller based on a gradient descent learning. This algorithm has a simple structure like conventional PID controller and a robustness to system parameters variation and different velocity command. To verify performances of the proposed adaptive PID controller, the speed control of nonlinear DC motor is performed. The simulation results show that the proposed control systems are effective in tracking a command velocity under system parameters variation.

• Journal of Biomedical Engineering Research
• /
• v.32 no.2
• /
• pp.93-104
• /
• 2011

The purpose of this study was to investigate lower-limb joint torque of the two groups as it changed by somatosensory stimulation during the descent down three stairs of different heights and to describe the difference between the two groups, which are young people group and elderly people group. Subjects of each groups climbed down a stair at four stimulation conditions, which are non-stimulation, tibialis anterior tendon stimulation, achilles tendon stimulation, tibialis anterior - achilles tendon stimulation. Motion capture data were collected using 3D optoelectric motion tracking system that utilizes active infrared LEDs, near infrared sensor and force plate. The obtained motion capture data was used to build 3D computer simulation model. The results show that lower-limb joint torque of the two groups changed with somatosensory stimulation as they descended the stairs and the joint torque of the two groups differed from each other.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.17 no.5
• /
• pp.598-603
• /
• 2007

To solve the rule explosion problem in multi-input fuzzy system, a method of converting a given fuzzy system to 2 layered hierarchical fuzzy system has been reported, where at the 1st layer, linearly independent fuzzy rule vectors generated from the given fuzzy system are used and, at the 2nd layer, linear combinations of these independent fuzzy rule vectors are used. In this paper, the steapest descent algorithm is presented to learn the fuzzy rule vectors and related coefficients for the equivalent 2 layered hierarchical structure. By simulation of learning of ball and beam control system, the feasibility of proposed learning scheme is shown.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.12 no.9
• /
• pp.1591-1598
• /
• 2008

Z. Cao had proposed NFRM(new fuzzy reasoning method) which infers in detail using relation matrix. In spite of the small inference rules, it shows good performance than mamdani's fuzzy inference method. In this paper, we propose Z. Cao's fuzzy inference method with learning ability which is used a gradient descent method in order to improve the performances. It is hard to determine the relation matrix elements by trial and error method. Because this method is needed many hours and effort. Simulation results are applied nonlinear systems show that the proposed inference method using a gradient descent method has good performances.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2017.05a
• /
• pp.392-395
• /
• 2017

본 연구에서는 일 증발접시 증발량 산정을 위한 딥러닝 (deep learning) 모형의 적용성을 평가하였다. 본 연구에서 적용된 딥러닝 모형은 deep belief network (DBN) 기반 deep neural network (DNN) (DBN-DNN) 모형이다. 모형 적용성 평가를 위하여 부산 관측소에서 측정된 기상자료를 활용하였으며, 증발량과의 상관성이 높은 기상변수들 (일사량, 일조시간, 평균지상온도, 최대기온)의 조합을 고려하여 입력변수집합 (Set 1, Set 2, Set 3)별 모형을 구축하였다. DBN-DNN 모형의 성능은 통계학적 모형성능 평가지표 (coefficient of efficiency, CE; coefficient of determination, $r^2$; root mean square error, RMSE; mean absolute error, MAE)를 이용하여 평가되었으며, 기존의 두가지 형태의 ANN (artificial neural network), 즉 모형학습 시 SGD (stochastic gradient descent) 및 GD (gradient descent)를 각각 적용한 ANN-SGD 및 ANN-GD 모형과 비교하였다. 효과적인 모형학습을 위하여 각 모형의 초매개변수들은 GA (genetic algorithm)를 이용하여 최적화하였다. 그 결과, Set 1에 대하여 ANN-GD1 모형, Set 2에 대하여 DBN-DNN2 모형, Set 3에 대하여 DBN-DNN3 모형이 가장 우수한 모형 성능을 나타내는 것으로 분석되었다. 비록 비교 모형들 사이의 모형성능이 큰 차이를 보이지는 않았으나, 모든 입력집합에 대하여 DBN-DNN3, DBN-DNN2, ANN-SGD3 순으로 모형 효율성이 우수한 것으로 나타났다.