• Journal of information and communication convergence engineering
• /
• 제19권1호
• /
• pp.48-53
• /
• 2021

Herein, we propose a document analysis system that analyzes papers or reports transformed into XML(Extensible Markup Language) format. It reads the document specified by the user, extracts keywords from the document, and compares the frequency of keywords to extract the top-three keywords. It maintains the order of the paragraphs containing the keywords and removes duplicated paragraphs. The frequency of the top-three keywords in the extracted paragraphs is re-verified, and the paragraphs are partitioned into 10 sections. Subsequently, the importance of the relevant areas is calculated and compared. By notifying the user of areas with the highest frequency and areas with higher importance than the average frequency, the user can read only the main content without reading all the contents. In addition, the number of paragraphs extracted through the deep learning model and the number of paragraphs in a section of high importance are predicted.

• 한국컴퓨터정보학회:학술대회논문집
• /
• 한국컴퓨터정보학회 2022년도 제66차 하계학술대회논문집 30권2호
• /
• 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.

• Structural Engineering and Mechanics
• /
• 제90권4호
• /
• pp.403-415
• /
• 2024

We looked at how the damping qualities of epoxy composites changed when different amounts of graphite nanoplatelets (GNP) were added, from 0% to 6% by weight. A mix of free and forced vibration tests helped us find the key GNP content that makes the damper ability better the most. We also created a Representative Volume Element (RVE) model to guess how the alloys would behave mechanically and checked these models against testing data. An Artificial Neural Network (ANN) was also used to guess how these compounds would react to motion. With proper hyperparameter tweaking, the ANN model showed good correlation (R²=0.98) with actual data, indicating its ability to predict complex material behavior. Combining these methods shows how GNPs impact epoxy composite mechanical properties and how machine learning might improve material design. We show how adding GNPs to epoxy composites may considerably reduce vibration. These materials may be used in industries that value vibration damping.

• Journal of information and communication convergence engineering
• /
• 제20권1호
• /
• pp.22-30
• /
• 2022

The reinforcement learning algorithm has proven its potential in solving sequential decision-making problems under uncertainties, such as finding paths to route data packets in wireless sensor networks. With reinforcement learning, the computation of the optimum path requires careful definition of the so-called reward function, which is defined as a linear function that aggregates multiple objective functions into a single objective to compute a numerical value (reward) to be maximized. In a typical defined linear reward function, the multiple objectives to be optimized are integrated in the form of a weighted sum with fixed weighting factors for all learning agents. This study proposes a reinforcement learning -based routing protocol for wireless sensor network, where different learning agents prioritize different objective goals by assigning weighting factors to the aggregated objectives of the reward function. We assign appropriate weighting factors to the objectives in the reward function of a sensor node according to its hop-count distance to the sink node. We expect this approach to enhance the effectiveness of multi-objective reinforcement learning for wireless sensor networks with a balanced trade-off among competing parameters. Furthermore, we propose SDN (Software Defined Networks) architecture with multiple controllers for constant network monitoring to allow learning agents to adapt according to the dynamics of the network conditions. Simulation results show that our proposed scheme enhances the performance of wireless sensor network under varied conditions, such as the node density and traffic intensity, with a good trade-off among competing performance metrics.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2002년도 ICCAS
• /
• pp.103.2-103
• /
• 2002

1. Introduction 2. Structure of BNN 3. Decision of weight value and threshold value 4. Principle of Extension in the ETL algorithm 5. Approximation problem of one circular region 6. Problem of synthetic image having four class 7. Conclusion

• 정보처리학회논문지B
• /
• 제9B권2호
• /
• pp.139-146
• /
• 2002

로봇 축구 시뮬레이션 게임은 하나의 동적 다중 에이전트 환경이다. 본 논문에서는 그러한 환경 하에서 각 에이전트의 동적 위치 결정을 위한 새로운 강화학습 방법을 제안한다. 강화학습은 한 에이전트가 환경으로부터 받는 간접적 지연 보상을 기초로 누적 보상값을 최대화할 수 있는 최적의 행동 전략을 학습하는 기계학습 방법이다. 따라서 강화학습은 입력-출력 쌍들이 훈련 예로 직접 제공되지 않는 다는 점에서 교사학습과 크게 다르다. 더욱이 Q-학습과 같은 비-모델 기반의 강화학습 알고리즘들은 주변 환경에 대한 어떤 모델도 학습하거나 미리 정의하는 것을 요구하지 않는다. 그럼에도 불구하고 이 알고리즘들은 에이전트가 모든 상태-행동 쌍들을 충분히 반복 경험할 수 있다면 최적의 행동전략에 수렴할 수 있다. 하지만 단순한 강화학습 방법들의 가장 큰 문제점은 너무 큰 상태 공간 때문에 보다 복잡한 환경들에 그대로 적용하기 어렵다는 것이다. 이런 문제점을 해결하기 위해 본 연구에서는 기존의 모듈화 Q-학습방법(MQL)을 개선한 적응적 중재에 기초한 모듈화 Q-학습 방법(AMMQL)을 제안한다. 종래의 단순한 모듈화 Q-학습 방법에서는 각 학습 모듈들의 결과를 결합하는 방식이 매우 단순하고 고정적이었으나 AMMQL학습 방법에서는 보상에 끼친 각 모듈의 기여도에 따라 모듈들에 서로 다른 가중치를 부여함으로써 보다 유연한 방식으로 각 모듈의 학습결과를 결합한다. 따라서 AMMQL 학습 방법은 큰 상태공간의 문제를 해결할 수 있을 뿐 아니라 동적인 환경변화에 보다 높은 적응성을 제공할 수 있다. 본 논문에서는 로봇 축구 에이전트의 동적 위치 결정을 위한 학습 방법으로 AMMQL 학습 방법을 사용하였고 이를 기초로 Cogitoniks 축구 에이전트 시스템을 구현하였다.

• 한국시뮬레이션학회논문지
• /
• 제22권1호
• /
• pp.53-61
• /
• 2013

본 논문에서는 시각정보와 경험이 가상환경에서 햅틱 디바이스로 상호작용 하는데 미치는 영향에 대한 연구이다. 이를 알아보기 위해 Load-on-task를 통하여 물체의 크기와 누적된 경험이 가상환경에서 무게를 예측하고 상호작용 하는데 영향을 미친다는 사실을 확인하였다. 또 가상환경에서 물체 이동 실험을 통해 물체에 대한 정보가 다를 경우 빠르게 잘못된 정보를 보상하고 안정적인 힘 조절을 확인하였다. 위 실험을 통해 가상환경에서의 경험이 물체의 시각적인 정보보다 많은 영향을 주지만, 경험이 없는 환경의 경우 시각적인 정보가 가상환경에서 물체와의 상호작용에 크게 영향을 주는 요소라는 것을 확인하였다. 따라서 가상환경에서 정밀한 조작(로봇 수술 등)이 필요한 의 경우 실제 물체의 물성치와 시각 데이터를 기반으로 유사한 환경을 조성해야 하며, 만약 그러한 환경 조성이 어려울 시에는 충분한 사전시험을 통하여 가상환경이 가진 특성을 경험할 수 있게 준비해야 한다.

• Journal of Nutrition and Health
• /
• 제16권2호
• /
• pp.81-88
• /
• 1983

This study was performed to investigate the influence of the amino acid composition of diet and environment on RNA, protein content in brain and learning ability in rats. Forty-two Sprague-Dawley male rats were divided into six groups according to type of diet, casein, soybean meal, or corn gluten and rearing condition, isolated or enriched. They were fed foods ad libitum for 6 weeks. A water maze was used to test behavioral performance for 3 weeks from 4th week. The rats were sacrificed at 6th week and their whole brains were taken and frozen for assay of RNA and protein. The results were summarized as follows : 1) The body weight gain for the experimental periods of corn gluten group was significantly lower than the casein and the soybean meal group. 2) The brain weight of the corn gluten group was significantly lower than the casein and the soybean meal group and the environmental enrichment slightly increased it among rats fed the corn gluten diet. 3) The total RNA contents were the greatest in the environmentally enriched casein group. The brain protein contents of the isolated corngluten group was the smallest. However, the contents of the enriched corn gluten group were similar to those of the others. 4) In the water maze test, the isolated corngluten group spent significantly more time than the others. Environmental enrichment could decrease time to perform the task of the maze.

• Structural Engineering and Mechanics
• /
• 제45권1호
• /
• pp.111-127
• /
• 2013

In this paper, a new version of gravitational search algorithm based on opposition-based learning (OBGSA) is introduced and applied for optimum design of reinforced concrete retaining walls. The new algorithm employs the opposition-based learning concept to generate initial population and updating agents' position during the optimization process. This algorithm is applied to minimize three objective functions include weight, cost and $CO_2$ emissions of retaining structure subjected to geotechnical and structural requirements. The optimization problem involves five geometric variables and three variables for reinforcement setups. The performance comparison of the new OBGSA and classical GSA algorithms on a suite of five well-known benchmark functions illustrate a faster convergence speed and better search ability of OBGSA for numerical optimization. In addition, the reliability and efficiency of the proposed algorithm for optimization of retaining structures are investigated by considering two design examples of retaining walls. The numerical experiments demonstrate that the new algorithm has high viability, accuracy and stability and significantly outperforms the original algorithm and some other methods in the literature.

• 전기전자학회논문지
• /
• 제22권1호
• /
• pp.14-20
• /
• 2018

The purpose of training a convolutional neural network (CNN) is to obtain weight factors that give high classification accuracies. The initial values of hyper-parameters affect the training results, and it is important to train a CNN with a suitable hyper-parameter set of a learning rate, a batch size, the initialization of weight factors, and an optimizer. We investigate the effects of a single hyper-parameter while others are fixed in order to obtain a hyper-parameter set that gives higher classification accuracies and requires shorter training time using a proposed VGG-like CNN for training since the VGG is widely used. The CNN is trained for four datasets of CIFAR10, CIFAR100, GTSRB and DSDL-DB. The effects of the normalization and the data transformation for datasets are also investigated, and a training scheme using merged datasets is proposed.