• 한국재난정보학회 논문집
• /
• 제19권1호
• /
• pp.44-59
• /
• 2023

연구목적: 대량전상자 발생 시 신속하고 정확한 환자분류가 진행되어야 최대한 많은 환자를 회복시켜 전장으로 돌려보낼 수 있다. 그러나 복잡한 전투현장에서 적은 의료인력으로 대량전상자의 환자분류를 시행하기란 임무는 과다하고 환경은 불확실하다. 따라서, 전투현장에서 의료인력을 보조하고 대체할 수 있는 인공지능 모델에 대해 논의하고자 한다. 연구방법:인공지능의 한 분야인 강화학습을 활용하여 환자분류 모델을 제시한다. 모델의 학습은 무작위로 설정된 환자의 상태와 병원시설의 의료능력을 고려하여 최대 다수의 환자가 치료받을 수 있는 정책을 찾도록 진행된다. 연구결과:강화학습 모델이 정상적으로 학습되었음은 누적 보상 값 등을 통하여 확인하였고, 학습된 모델이 정확하게 환자를 분류하는 것은 생존자 수를 통해 확인하였다. 또한, 규칙 기반 모델과 비교하여 성능을 검증하였으며, 강화학습 모델이 규칙 기반 모델에 비해 약 10%만큼 더 많은 환자를 생존시킬 수 있었다. 결론:강화학습을 이용한 환자분류 모델은 의료인력의 대량전상자 환자분류 의사결정을 보조하고 대체하는 대안으로 활용 가능하다.

• Computers and Concrete
• /
• 제32권3호
• /
• pp.327-337
• /
• 2023

Concrete is the most widely used building material, with various types including high- and ultra-high-strength, reinforced, normal, and lightweight concretes. However, accurately predicting concrete properties is challenging due to the geotechnical design code's requirement for specific characteristics. To overcome this issue, researchers have turned to new technologies like machine learning to develop proper methodologies for concrete specification. In this study, we propose a highly accurate deep learning-based predictive model to investigate the compressive strength (UCS) of lightweight concrete with natural aggregates (pumice). Our model was implemented on a database containing 249 experimental records and revealed that water, cement, water-cement ratio, fine-coarse aggregate, aggregate substitution rate, fine aggregate replacement, and superplasticizer are the most influential covariates on UCS. To validate our model, we trained and tested it on random subsets of the database, and its performance was evaluated using a confusion matrix and receiver operating characteristic (ROC) overall accuracy. The proposed model was compared with widely known machine learning methods such as MLP, SVM, and DT classifiers to assess its capability. In addition, the model was tested on 25 laboratory UCS tests to evaluate its predictability. Our findings showed that the proposed model achieved the highest accuracy (accuracy=0.97, precision=0.97) and the lowest error rate with a high learning rate (R2=0.914), as confirmed by ROC (AUC=0.971), which is higher than other classifiers. Therefore, the proposed method demonstrates a high level of performance and capability for UCS predictions.

• 한국멀티미디어학회논문지
• /
• 제21권2호
• /
• pp.98-107
• /
• 2018

In this paper, we propose a method to classify the skeletal maturity with a small amount of hand wrist X-ray image using deep learning-based meta-learning. General deep-learning techniques require large amounts of data, but in many cases, these data sets are not available for practical application. Lack of learning data is usually solved through transfer learning using pre-trained models with large data sets. However, transfer learning performance may be degraded due to over fitting for unknown new task with small data, which results in poor generalization capability. In addition, medical images require high cost resources such as a professional manpower and mcuh time to obtain labeled data. Therefore, in this paper, we use meta-learning that can classify using only a small amount of new data by pre-trained models trained with various learning tasks. First, we train the meta-model by using a separate data set composed of various learning tasks. The network learns to classify the bone maturity using the bone maturity data composed of the radiographs of the wrist. Then, we compare the results of the classification using the conventional learning algorithm with the results of the meta learning by the same number of learning data sets.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1997년도 한국자동제어학술회의논문집; 한국전력공사 서울연수원; 17-18 Oct. 1997
• /
• pp.215-218
• /
• 1997

A dynamic back-propagation neural network is addressed for adaptive neural control system to approximate non-linear control system rather than static networks. It has the capability to represent the approximation of nonlinear system without mathematical analysis and to carry out the on-line learning algorithm for real time application. The simulated results show fast tracking capability and adaptive response by using dynamic back-propagation neurons.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 2001년도 ICCAS
• /
• pp.146.2-146
• /
• 2001

Recently, a need of non-industrial robot, such as service, medical, entertainment and house-keeping robot, has been increased. Therefore, the capability of robot which can do intelligent behavior like interaction with men and its environment become more prominent than the capability of executing simple repetitive task. To implement an intelligent robot which provides intelligent behavior, an effective system architecture including perception, learning, reasoning and action part is necessary. Control architectures for intelligent robot can be divided into two different classes. One is deliberative type controller which is applicate to high level intelligence like environment ...

• 대한전기학회:학술대회논문집
• /
• 대한전기학회 2002년도 하계학술대회 논문집 D
• /
• pp.2143-2145
• /
• 2002

This paper proposes the control problem of an inverted pendulum system based on adaptive fuzzy sliding mode. The universal approximating capability, learning ability, adaptation capability and disturbance rejection are collected in one control strategy. The proposed scheme does not require an accurate dynamic model and the joint acceleration measurement, yet it guarantees asymptotic trajectory tracking. Experimental results perform with an inverted pendulum to show the effectiveness of the approach.

• 전자공학회논문지B
• /
• 제32B권11호
• /
• pp.1527-1535
• /
• 1995

This paper introduces an isolated word recognition system realized on a neurocomputer called E-MIND II, which is a 2-D torus wavefront array processor consisting of 256 DNP IIs. The DNP II is an all digital VLSI unit processor for the EMIND II featuring the emulation capability of more than thousands of neurons, the 40 MHz clock speed, and the on-chip learning. Built by these PEs in 2-D toroidal mesh architecture, the E- MIND II can be accelerated over 2 Gcps computation speed. In this light, the advantages of the E-MIND II in its capability of computing speed, scalability, computer interface, and learning are especially suitable for real time application such as speech recognition. We show how to map a TDNN structure on this array and how to code the learning and recognition algorithms for a user independent isolated word recognition. Through hardware simulation, we show that recognition rate of this system is about 97% for 30 command words for a robot control.

• 융합경영연구
• /
• 제8권1호
• /
• pp.36-49
• /
• 2020

Purpose - This study investigates factors that affected the utilization of trade support policies and further analyzed how the utilization of trade support policies affected export performance. Research design, data, and methodology - With a sample of 223 small and medium-sized export firms from South Korea, this study examines the determinants of the utilization level of trade support system such as export market orientation, learning orientation, network capability and environmental uncertainty by regression analysis. Results - Export market orientation have a positive effect on the utilization of the trade support system and there is positive relationship between learning orientation and the utilization of trade support system. And network capabilities have had a positive impact on the utilization of the trade support system but there is no relationship between environmental uncertainty and the utilization of trade support system. The utilization of the trade support system had a positive effect on export performance. Conclusions - The internal and external factors of the organization have affected small and medium-sized export firms use of trade support systems. The utilization of trade support system can enhance positive export performance by providing valuable information and resource to external knowledge and also to complementary resources from the external partners.

• International journal of advanced smart convergence
• /
• 제8권4호
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.

• 전자통신동향분석
• /
• 제33권5호
• /
• pp.13-23
• /
• 2018

Deep learning is a promising solution to a number of complex problems based on its inherent capability to approximate almost all types of functions without the demand for handcrafted feature extraction. New wireless transmission and access schemes based on deep learning are being increasingly proposed as substitutes for existing approaches, providing a lower complexity and better performance gain. Among such schemes, a communications system is viewed as an end-to-end autoencoder. The learning process applied in autoencoders can automatically deal with some nonlinear or unknown properties in communications systems. Deep learning can also be used to optimize each processing block for required tasks such as channel decoding, signal detection, and multiple access. On top of recent related research trends, we suggest appropriate research approaches for communications systems to adopt deep learning.