• International Journal of Control, Automation, and Systems
• /
• v.4 no.6
• /
• pp.725-735
• /
• 2006

This paper deals with the problem of self-learning cooperative motion control for the pushing task of a humanoid robot in the sagittal plane. A model with 27 linked rigid bodies is developed to simulate the system dynamics. A simple genetic algorithm(SGA) is used to find the cooperative motion, which is to minimize the total energy consumption for the entire humanoid robot body. And the multi-layer neural network based on backpropagation(BP) is also constructed and applied to generalize parameters, which are obtained from the optimization procedure by SGA, in order to control the system.

• THE INTERNATIONAL COMMERCE & LAW REVIEW
• /
• v.52
• /
• pp.3-22
• /
• 2011

This study examines the seller's liability to transfer the property to the buyer. Even though contracting parties choose CISG as the governing law regulating their obligations and rights by means of their contract, CISG does not concern with the effect generated by the transfer of property. Thus, the issues of the property is settled in conformity with the domestic law applicable by virtue of the rules of private international law. By considering the general rules of the transfer of property in goods under SGA and KCC as the lex rei sitae, the difference of requirements to pass the property between them is analyzed and then the reasons why the transfer of property is importantly considered are discussed. In addition, as CISG does not exclude completely the matters concerning the property and provides the provision like Art 41, the seller's liability to deliver goods free from the third party right or claim is examined under Art 41. Lastly, the practical advice is suggested.

• Proceedings of the Korean Institute of Intelligent Systems Conference
• /
• 2007.04a
• /
• pp.33-36
• /
• 2007

최근에 인공지능 분야에서는, 국내외적으로 강화학습(reinforcement learning)에 관한 관심이 크게 증폭되고 있다. 강화학습의 최근 경향을 살펴보면, 크게 가치함수를 직접 활용하는 방법(value function-based methods), 제어 전략에 대한 탐색을 활용하는 방법(policy search methods), 그리고 액터-크리틱 방법(actor-critic methods)의 세가지 방향으로 발전하고 있음을 알 수 있다. 본 논문에서는 이중 세 번째 부류인 액터-크리틱 방법 중 NAC(natural actor-critic) 기법의 한 종류인 RLS-NAC(recursive least-squares based natural actor-critic) 알고리즘을 다양한 트레이스 감쇠계수를 사용하여 연속제어입력(real-valued control inputs)으로 제어되는 Kimura의 기는 로봇에 대해 적용해보고, 그 성능을 기존의 SGA(stochastic gradient ascent) 알고리즘을 이용하여 학습한 경우와 비교해보도록 한다.

• Transactions of the Korean Society of Mechanical Engineers A
• /
• v.27 no.1
• /
• pp.158-166
• /
• 2003

The paper describes the development and application of advanced evolutionary computing techniques referred to as micro genetic algorithms ($\mu$GA) in the context of engineering design optimization. The basic concept behind $\mu$GA draws from the use of small size of population irrespective of the bit string length in the representation of design variable. Such strategies also demonstrate the faster convergence capability and more savings in computational resource requirements than simple genetic algorithms (SGA). The paper first explores ten-bar truss design problems to see the optimization performance between $\mu$GA and SGA. Subsequently, $\mu$GA is applied to a realistic engineering design problem in the injection molding process optimization.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.35C no.2
• /
• pp.48-55
• /
• 1998

Structural learning methods of MLP classifiers for a given application using genetic algorithms have been studied. In the methods, however, the search space for an optimal structure is increased exponentially for the physical application of high diemension-multi calss. In this paperwe propose a method of MLP classifiers using species genetic algorithm(SGA), a modified GA. In SGA, total search space is divided into several subspaces according to the number of hidden units. Each of the subdivided spaces is called "species". We eliminate low promising species from the evoluationary process in order to reduce the search space. experimental results show that the proposed method is more efficient than the conventional genetic algorithm methods in the aspect of the misclassification ratio, the learning rate, and the structure.structure.

• Asian Pacific Journal of Cancer Prevention
• /
• v.16 no.18
• /
• pp.8173-8176
• /
• 2016

Cancer patients frequently experience malnutrition. Cancer and cancer therapy effects nutritional status through alterations in the metabolic system and reduction in food intake. In the present study, fifty seven cancer patients were selected as subjects from the oncology ward of Cachar Cancer Hospital and Research Centre, Silchar, India. Evaluation of nutritional status of cancer patients during treatment was carried out by scored Patient-Generated Subjective Global Assessment (PG-SGA). The findings of PG-SGA showed that 15.8% (9) were well nourished, 31.6% (18) were moderately or suspected of being malnourished and 52.6% (30) were severely malnourished. The prevalence of malnutrition was highest in lip/oral (33.33%) cancer patients. The study showed that the prevalence of malnutrition (84.2%) was high in cancer patients during treatment.

• Journal of Korean Society of Industrial and Systems Engineering
• /
• v.23 no.59
• /
• pp.11-20
• /
• 2000

The Job Shop Scheduling Problem(JSSP) is one of the most general and difficult of all traditional scheduling problems. The goal of this research is to develop an efficient scheduling method based on single genetic algorithm(SGA) and parallel genetic algorithm (PGA) to address JSSP. In this scheduling method, new genetic operator, generating method of initial population are developed and island model PGA are proposed. The scheduling method based on PGA are tested on standard benchmark JSSP. The results were compared with SGA and another GA-based scheduling method. The PGA search the better solution or improves average of solution in benchmark JSSP. Compared to traditional GA, the proposed approach yields significant improvement at a solution.

• Proceedings of the KIEE Conference
• /
• 1998.11b
• /
• pp.627-629
• /
• 1998

We propose a robust blind equalization algorithm based on quadrant-partitioned constellations for QAM demodulation. The algorithm divides the received M-QAM constellations into simple four quadrant. The channel equalization for symbols in each quadrant can be accomplished fast and reliably using the Constant Modulus Algorithm(CMA) and the Stop-and-Go Algorithm(SGA). Test results confirm that the proposed algorithm with lower complexity outperforms both the CMA and the SGA in reducing the SER as well as the MSE at the equalizer output.

• Proceedings of the KIEE Conference
• /
• 2001.07d
• /
• pp.1996-1998
• /
• 2001

This paper presents the closed-loop composite control for weakly coupled bilinear systems with a quadratic performance criterion. The Riccati equation for weakly coupled bilinear system is decomposed into three reduced Riccati equations by the weak coupling theory, and we obtain optimal solutions of each reduced Riccati equation using successive Galerkin approximation(SGA). We design the composite control law that consists of optimal solutions of each reduced Riccati equation. The proposed algorithm reduces the disadvantages of SGA method.

• Clinical and Experimental Pediatrics
• /
• v.61 no.4
• /
• pp.114-120
• /
• 2018

Purpose: Routine screening for toxoplasmosis, rubella, cytomegalovirus (CMV), and herpes simplex virus (TORCH) in intrauterine growth restriction (IUGR) and small for gestational age (SGA) neonates has become a common practice. However, the incidence of TORCH varies across countries, and the cost of TORCH testing may be disadvantageous compared to disease-specific screening. To evaluate the efficacy of TORCH screening, the medical charts of IUGR or SGA neonates born in a single institution in Bucheon, Korea from 2011 to 2015 were reviewed. Methods: The clinical data of the 126 IUGR or SGA neonates were gathered, including gestational age, Apgar scores, neonatal sonographic findings, chromosome study, morbidities, developmental follow-up, and growth catch-up. Maternal factors including underlying maternal disease and fetal sonography were collected, and placental findings were recorded when available. TORCH screening was done using serum IgM, CMV urine culture, quantification of CMV DNA with real-time polymerase chain reaction, and rapid plasma reagin qualitative test for syphilis. Tests were repeated only for those with positive results. Results: Of the 119 TORCH screenings, only one was positive for toxoplasmosis IgM. This result was deemed false positive due to negative IgM on repeated testing and the absence of clinical symptoms. Conclusion: Considering the incidence and risk of TORCH in Korea, the financial burden of TORCH screening, and the single positive TORCH finding in our study, we suggest disease-specific screening based on maternal history and the clinical symptoms of the neonate. Regarding CMV, which may present asymptomatically, universal screening may be appropriate upon cost-benefit analysis.