• 한국컴퓨터정보학회논문지
• /
• 제27권11호
• /
• pp.13-18
• /
• 2022

본 논문에서는 학습에 참여하는 각 디바이스의 모델들로부터 성능검증에 따라 가중치를 두어 글로벌 모델을 업데이트하는 VW-FedAVG(Validation based Weighted FedAVG)를 두 가지 방식으로 제안 한다. 첫 번째 방식은 서버 검증(Server side Validation) 구조로 글로벌 모델을 업데이트 하기 전에 각 로컬 클라이언트 모델을 하나의 전체 검증 데이터셋을 통해 검증하도록 설계 했다. 두 번째는 클라이언트 검증(Client side Validation) 구조로 검증 데이터셋을 각 클라이언트에 고르게 분배하여 검증을 한 후 글로벌 모델을 업데이트 하는 방식으로 설계 했다. 전체 실험에 적용한 데이터셋은 MNIST, CIFAR-10으로 이미지 분류에 대해 IID, Non-IID 분포에서 기존 연구 대비 더 높은 정확도를 얻을 수 있었다.

• 한국시뮬레이션학회논문지
• /
• 제30권4호
• /
• pp.1-8
• /
• 2021

복잡하고 비선형적인 특징을 갖는 시계열 데이터를 예측하기 위해 딥러닝 기법이 널리 사용되고 있다. 본 연구에서는 최근에 개발된 WaveNet을 개선하고 워크포워드 검증 기법을 적용하여 전력 소비량 데이터를 24시간 이전에 예측하고자 한다. 원래 WaveNet은 오디오 데이터 예측에 사용하고자 고안되었으며, 장기간의 데이터를 효과적으로 예측하기 위해 1차원 팽창인과 합성곱(1D dilated causal convolution)을 사용한다. 먼저, WaveNet이 부호화된 정수 값이 아니라 실수 값을 출력하여 전력 데이터를 예측하기 적합하도록 개선하였다. 다음으로 학습 과정에 적용된 하이퍼파라미터(입력 기간, 배치 크기, WaveNet 블록 개수, 팽창 비율, 학습률 변경)를 조정하여 적절한 성능을 나타내도록 하였다. 마지막으로 성능 평가를 통해 전통적인 홀드아웃 검증 기법보다 본 연구에서 사용한 워크포워드 검증 기법이 전력 소비량 데이터 예측에 우수함 성능을 나타냄을 확인하였다.

• 융합신호처리학회논문지
• /
• 제20권3호
• /
• pp.132-137
• /
• 2019

의료영상으로 생성된 데이터의 양은 전문적인 시각적 분석 한계를 점점 초과하여, 자동화된 의료영상 분석의 필요성이 증가되고 있는 실정이다. 이러한 이유 등으로 인하여 본 논문에서는 정상소견과 종양소견을 보이는 각각의 뇌 실질 MRI 의료영상을 이용하여 Inception V3 딥러닝 모델을 이용한 종양 유무에 따른 분류 및 정확도를 평가하였다. 연구 결과, 딥러닝 모델의 정확도 평가는 학습 데이터 세트의 경우 90%, 검증 데이터 세트의 경우 86%의 정확도를 나타내었다. 손실률 평가에서는 학습 데이터 세트의 경우 0.56, 검증 데이터 세트의 경우 1.28의 손실률을 나타내었다. 향 후 연구에서는 딥러닝 모델의 성능 향상 및 평가의 신뢰성 확보를 위하여 공개된 의료영상의 데이터를 충분히 확보하고, 라벨링 분류 작업을 통한 라벨링의 정확도를 개선하여 모델링을 구현해 볼 필요가 있다고 사료된다.

• Computers and Concrete
• /
• 제19권6호
• /
• pp.717-724
• /
• 2017

In this paper, four formulas are proposed via gene expression programming (GEP)-based models and regression analysis (RA) to predict the flexural strength ($f_s$) values of mortars containing different mineral admixtures that are ground granulated blast-furnace slag (GGBFS), silica fume (SF) and fly ash (FA) at different ages. Three formulas obtained from the GEP-I, GEP-II and GEP-III models are constituted to predict the $f_s$ values from the age of specimen, water-binder ratio and compressive strength. Besides, one formula obtained from the RA is constituted to predict the $f_s$ values from the compressive strength. To achieve these formulas in the GEP and RA models, 972 data of the experimental studies presented with mortar mixtures were gathered from the literatures. 734 data of the experimental studies are divided without pre-planned for these formulas achieved from the training and testing sets of GEP and RA models. Beside, these formulas are validated with 238 data of experimental studies un-employed in training and testing sets. The $f_s$ results obtained from the training, testing and validation sets of these formulas are compared with the results obtained from the experimental studies and the formulas given in the literature for concrete. These comparisons show that the results of the formulas obtained from the GEP and RA models appear to well compatible with the experimental results and find to be very credible according to the results of other formulas.

• Structural Engineering and Mechanics
• /
• 제46권5호
• /
• pp.673-693
• /
• 2013

This paper presents a systematic design and training procedure for the feed-forward back-propagation neural network (NN) modeling of both forward and inverse behavior of a rotary magnetorheological (MR) damper based on experimental data. For the forward damper model, with damper force as output, an optimization procedure demonstrates accurate training of the NN architecture with only current and velocity as input states. For the inverse damper model, with current as output, the absolute value of velocity and force are used as input states to avoid negative current spikes when tracking a desired damper force. The forward and inverse damper models are trained and validated experimentally, combining a limited number of harmonic displacement records, and constant and half-sinusoidal current records. In general the validation shows accurate results for both forward and inverse damper models, where the observed modeling errors for the inverse model can be related to knocking effects in the measured force due to the bearing plays between hydraulic piston and MR damper rod. Finally, the validated models are used to emulate pure viscous damping. Comparison of numerical and experimental results demonstrates good agreement in the post-yield region of the MR damper, while the main error of the inverse NN occurs in the pre-yield region where the inverse NN overestimates the current to track the desired viscous force.

• 한국정밀공학회:학술대회논문집
• /
• 한국정밀공학회 1996년도 추계학술대회 논문집
• /
• pp.384-388
• /
• 1996

The closed-loop system modeling of an Active/semiactive suspension system has been accomplished through an artificial neural Networks. The 7DOF full model as the system equation of motion has been derived and the output feedback linear quadratic regulator has been designed for the control purpose. For the neural networks training set of a sample data has been obtained through the computer simulation. A 7DOF full model with LQR controller simulated under the several road conditions such as sinusoidal bumps and the rectangular bumps. A general multilayer perceptron neural network is used for the dynamic modeling and the target outputs are feedback to the input layer. The Backpropagation method is used as the training algorithm. The modeling of system and the model validation have been shown through computer simulations.

• 한국병원경영학회지
• /
• 제8권1호
• /
• pp.81-94
• /
• 2003

This paper aims at developing a prediction model and analyzing a sensitivity for the outpatient's overall satisfaction on utilizing hospital services by using data mining techniques within the context of customer satisfaction. From a total of 900 outpatient cases, 80 percent were randomly selected as the training group and the other 20 percent as the validation group. Cases in the training group were used in the development of the CHAID and Neural Networks. The validation group was used to test the performance of these models. The major findings may be summarized as follows: the CHAID provided six useful predictors - satisfaction with treatment level, satisfaction with healthcare facilities and equipments, satisfaction with registration service, awareness of hospital reputation, satisfaction with staffs courtesy and responsiveness, and satisfaction with nurses kindness. The prediction accuracy rates based on MLP (77.90%) is superior to RBF (76.80%).

• International Journal of Computer Science & Network Security
• /
• 제21권12spc호
• /
• pp.526-538
• /
• 2021

Machine and deep learning-based models are emerging techniques that are being used to address prediction problems in biomedical data analysis. DNA sequence prediction is a critical problem that has attracted a great deal of attention in the biomedical domain. Machine and deep learning-based models have been shown to provide more accurate results when compared to conventional regression-based models. The prediction of the gene sequence that leads to cancerous diseases, such as prostate cancer, is crucial. Identifying the most important features in a gene sequence is a challenging task. Extracting the components of the gene sequence that can provide an insight into the types of mutation in the gene is of great importance as it will lead to effective drug design and the promotion of the new concept of personalised medicine. In this work, we extracted the exons in the prostate gene sequences that were used in the experiment. We built a Deep Neural Network (DNN) and Bi-directional Long-Short Term Memory (Bi-LSTM) model using a k-mer encoding for the DNA sequence and one-hot encoding for the class label. The models were evaluated using different classification metrics. Our experimental results show that DNN model prediction offers a training accuracy of 99 percent and validation accuracy of 96 percent. The bi-LSTM model also has a training accuracy of 95 percent and validation accuracy of 91 percent.

• 대한건축학회논문집:구조계
• /
• 제36권3호
• /
• pp.113-120
• /
• 2020

The purpose of this study is to compare the models of Deep Learning-based Convolution Neural Network(CNN) for concrete crack detection. The comparison models are AlexNet, GoogLeNet, VGG16, VGG19, ResNet-18, ResNet-50, ResNet-101, and SqueezeNet which won ImageNet Large Scale Visual Recognition Challenge(ILSVRC). To train, validate and test these models, we constructed 3000 training data and 12000 validation data with 256×256 pixel resolution consisting of cracked and non-cracked images, and constructed 5 test data with 4160×3120 pixel resolution consisting of concrete images with crack. In order to increase the efficiency of the training, transfer learning was performed by taking the weight from the pre-trained network supported by MATLAB. From the trained network, the validation data is classified into crack image and non-crack image, yielding True Positive (TP), True Negative (TN), False Positive (FP), False Negative (FN), and 6 performance indicators, False Negative Rate (FNR), False Positive Rate (FPR), Error Rate, Recall, Precision, Accuracy were calculated. The test image was scanned twice with a sliding window of 256×256 pixel resolution to classify the cracks, resulting in a crack map. From the comparison of the performance indicators and the crack map, it was concluded that VGG16 and VGG19 were the most suitable for detecting concrete cracks.

• Nuclear Engineering and Technology
• /
• 제54권8호
• /
• pp.3027-3033
• /
• 2022

Background: In this study, various types of deep-learning models for predicting in vitro radiosensitivity from gene-expression profiling were compared. Methods: The clonogenic surviving fractions at 2 Gy from previous publications and microarray gene-expression data from the National Cancer Institute-60 cell lines were used to measure the radiosensitivity. Seven different prediction models including three distinct multi-layered perceptrons (MLP), four different convolutional neural networks (CNN) were compared. Folded cross-validation was applied to train and evaluate model performance. The criteria for correct prediction were absolute error < 0.02 or relative error < 10%. The models were compared in terms of prediction accuracy, training time per epoch, training fluctuations, and required calculation resources. Results: The strength of MLP-based models was their fast initial convergence and short training time per epoch. They represented significantly different prediction accuracy depending on the model configuration. The CNN-based models showed relatively high prediction accuracy, low training fluctuations, and a relatively small increase in the memory requirement as the model deepens. Conclusion: Our findings suggest that a CNN-based model with moderate depth would be appropriate when the prediction accuracy is important, and a shallow MLP-based model can be recommended when either the training resources or time are limited.