• Bulletin of the Korean Chemical Society
• /
• v.24 no.6
• /
• pp.744-750
• /
• 2003

In quantum dynamics of many-body systems, to identify the Hamiltonian becomes more difficult very rapidly as the number of degrees of freedom increases. In order to simplify the dynamics and to deduce dynamically relevant Hamiltonian information, it is desirable to control the dynamics to lie within a reduced space. With a judicious choice for the cost functional, the closed loop optimal control experiments can be manipulated efficiently to steer the dynamics to lie within a subspace of the system eigenstates without requiring any prior detailed knowledge about the system Hamiltonian. The procedure is simulated for optimally controlled population transfer experiments in the system of two degrees of freedom. To show the feasibility of steering the dynamics to lie in a specified subspace, the learning algorithms guiding the dynamics are presented along with frequency filtering. The results demonstrate that the optimal control fields derive the system to the desired target state through the desired subspace.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.34C no.1
• /
• pp.42-50
• /
• 1997

Vector quantization has shown to be useful for compressing data related with a wide rnage of applications such as image processing, speech processing, and weather satellite. Neural networks of images this paper propses a efficient neural network learning algorithm, called learning count control algorithm based on the frquency sensitive learning algorithm. This algorithm can train a results more codewords can be assigned to the sensitive region of the human visual system and the quality of the reconstructed imate can be improved. We use a human visual systrem model that is a cascade of a nonlinear intensity mapping function and a modulation transfer function with a bandpass characteristic.

• International Journal of Internet, Broadcasting and Communication
• /
• v.14 no.2
• /
• pp.119-128
• /
• 2022

Recently, due to the development of related technologies for autonomous vehicles, driving work is changing more safely. However, the development of support technologies for level 5 full autonomous driving is still insufficient. That is, even in the case of an autonomous vehicle, the driver needs to drive through forward attention while driving. In this paper, we propose a method to monitor driving tasks by recognizing driver behavior. The proposed method uses pre-trained deep convolutional neural network models to recognize whether the driver's face or body has unnecessary movement. The use of pre-trained Deep Convolitional Neural Network (DCNN) models enables high accuracy in relatively short time, and has the advantage of overcoming limitations in collecting a small number of driver behavior learning data. The proposed method can be applied to an intelligent vehicle safety driving support system, such as driver drowsy driving detection and abnormal driving detection.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2022.07a
• /
• pp.67-70
• /
• 2022

NLP 분야에서 적극 활용되기 시작한 Transformer 모델을 Vision 분야에서 적용하기 시작하면서 object detection과 segmentation 등 각종 분야에서 기존 CNN 기반 모델의 정체된 성능을 극복하며 향상되고 있다. 또한, label 데이터 없이 이미지들로만 자기지도학습을 한 ViT(Vision Transformer) 모델을 통해 이미지에 포함된 여러 중요한 객체의 영역을 검출하는 saliency map을 추출할 수 있게 되었으며, 이로 인해 ViT의 자기지도학습을 통한 object detection과 semantic segmentation 연구가 활발히 진행되고 있다. 본 논문에서는 ViT 모델 뒤에 classifier를 붙인 모델에 일반 학습한 모델과 자기지도학습의 pretrained weight을 사용해서 전이학습한 모델의 시각화를 통해 각 saliency map들을 비교 분석하였다. 이를 통해, 클래스 분류 학습 기반 전이학습이 transformer의 saliency map에 미치는 영향을 확인할 수 있었다.

• Annual Conference on Human and Language Technology
• /
• 2022.10a
• /
• pp.320-323
• /
• 2022

상호참조해결은 문서에서 명사, 대명사, 명사구 등의 멘션 후보를 식별하고 동일한 개체를 의미하는 멘션들을 찾아 그룹화하는 태스크이다. 딥러닝 기반의 한국어 상호참조해결 연구들에서는 BERT를 이용하여 단어의 문맥 표현을 얻은 후 멘션 탐지와 상호참조해결을 동시에 수행하는 End-to-End 모델이 주로 연구가 되었으며, 최근에는 스팬 표현을 사용하지 않고 시작과 끝 표현식을 통해 상호참조해결을 빠르게 수행하는 Start-to-End 방식의 한국어 상호참조해결 모델이 연구되었다. 최근에 한국어 상호참조해결을 위해 구축된 ETRI 데이터셋은 WIKI, QA, CONVERSATION 등 다양한 도메인으로 이루어져 있으며, 신규 도메인의 데이터가 추가될 경우 신규 데이터가 추가된 전체 학습데이터로 모델을 다시 학습해야 하며, 이때 많은 시간이 걸리는 문제가 있다. 본 논문에서는 이러한 상호참조해결 모델의 도메인 적응에 Continual learning을 적용해 각기 다른 도메인의 데이터로 모델을 학습 시킬 때 이전에 학습했던 정보를 망각하는 Catastrophic forgetting 현상을 억제할 수 있음을 보인다. 또한, Continual learning의 성능 향상을 위해 2가지 Transfer Techniques을 함께 적용한 실험을 진행한다. 실험 결과, 본 논문에서 제안한 모델이 베이스라인 모델보다 개발 셋에서 3.6%p, 테스트 셋에서 2.1%p의 성능 향상을 보였다.

• Educational Technology International
• /
• v.6 no.2
• /
• pp.69-85
• /
• 2005

The purpose of the article is to address a new instructional approach to a complex engineering course. We design a novel instructional method that combines mobile technology, simulation program, collaborative teamwork, problem-solving process, and a variety of evaluation techniques. We suggested five instructional principles that might be required to change the fundamental educational process by which learning is done. The proposed instructional method is expected to aspire for new perspectives on complex learning environment. Nevertheless we solely began by the research on the development of students' complex problem-solving performance in a complex engineering course, the new instructional method in the article can promote the adoption of new instructional methods and strategies across different knowledge domains. In addition, the instructional method can provide a valuable bridge to acquisition and transfer of problem solving, motivation, and meaning learning.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.3 no.2
• /
• pp.134-146
• /
• 2009

Traffic classification seeks to assign packet flows to an appropriate quality of service(QoS) class based on flow statistics without the need to examine packet payloads. Classification proceeds in two steps. Classification rules are first built by analyzing traffic traces, and then the classification rules are evaluated using test data. In this paper, we use self-organizing map and K-means clustering as unsupervised machine learning methods to identify the inherent classes in traffic traces. Three clusters were discovered, corresponding to transactional, bulk data transfer, and interactive applications. The K-nearest neighbor classifier was found to be highly accurate for the traffic data and significantly better compared to a minimum mean distance classifier.

• Computers and Concrete
• /
• v.32 no.6
• /
• pp.625-637
• /
• 2023

The application of ML approaches in determining the resisting capacity of fire damaged RC columns is introduced in this paper, on the basis of analysis data driven ML modeling. Considering the characteristics of the structural behavior of fire damaged RC columns, the representative five approaches of Kernel SVM, ANN, RF, XGB and LGBM are adopted and applied. Additional partial monotonic constraints are adopted in modelling, to ensure the monotone decrease of resisting capacity in RC column with fire exposure time. Furthermore, additional suggestions are also added to mitigate the heterogeneous composition of the training data. Since the use of ML approaches will significantly reduce the computation time in determining the resisting capacity of fire damaged RC columns, which requires many complex solution procedures from the heat transfer analysis to the rigorous nonlinear analyses and their repetition with time, the introduced ML approach can more effectively be used in large complex structures with many RC members. Because of the very small amount of experimental data, the training data are analytically determined from a heat transfer analysis and a subsequent nonlinear finite element (FE) analysis, and their accuracy was previously verified through a correlation study between the numerical results and experimental data. The results obtained from the application of ML approaches show that the resisting capacity of fire damaged RC columns can effectively be predicted by ML approaches.

• The Korean Journal of Emergency Medical Services
• /
• v.15 no.2
• /
• pp.55-66
• /
• 2011

Purpose: The purpose of this study was to develop a scenario and evaluate the performance of paramedic students in head trauma simulation education. Method: This study selected a refined head trauma scenario that was developed by graduate students during class from september to November, 2010. Evaluation on implementation of head trauma simulation was conducted on seventeen paramedic students divided into four groups during November, 2010. Results: 1. The head trauma scenario was developed according to the patient assessment for approximately 10 minutes. It contained scene size-up, initial assessment and intervention, rapid trauma assessment and intervention, and decision of transfer. 2. The average time turned out to be 9 min and 36 sec after simulation learning. Total mean score in simulation performance was 2.20, the highest score was 2.44 in initial assessment and intervention, and the lowest score was 1.5 in decision of transfer. 3. Confidence mean was high with the score of 1.0. Conclusion: The finding of this study demonstrate that the simulation education can improve problem-solving ability and critical thinking, and increase the confidence in prehospital emergency care; therefore, simulation may be the new effective paramedic education strategy and simulation learning is needed for further development of various scenarios.

• Journal of Korean Institute of Industrial Engineers
• /
• v.43 no.3
• /
• pp.192-202
• /
• 2017

This study develops a text spam filtering system for Facebook based on two variable categories: keywords learned from Instagram and meta-information of Facebook posts. Since there is no explicit labels for spam/ham posts, we utilize hash tags in Instagram to train classification models. In addition, the filtering accuracy is enhanced by considering meta-information of Facebook posts. To verify the proposed filtering system, we conduct an empirical experiment based on a total of 1,795,067 and 761,861 Facebook and Instagram documents, respectively. Employing random forest as a base classification algorithm, experimental result shows that the proposed filtering system yield 99% and 98% in terms of filtering accuracy and F1-measure, respectively. We expect that the proposed filtering scheme can be applied other web services suffering from massive spam posts but no explicit spam labels are available.