• 한국산업정보학회논문지
• /
• 제22권5호
• /
• pp.17-24
• /
• 2017

최근 딥러닝, 즉 거대 또는 깊은 인공신경망을 사용한 기술이 놀라운 성능을 보이고 있고, 점차로 그 네트워크의 규모가 커지고 있다. 하지만, 신경망 크기의 증가는 계산양의 증가로 이어져서 회로의 복잡성, 가격, 발열, 실시간성 제약 등의 문제를 야기한다. 또한, 신경망 연결에는 많은 중복성이 존재한다, 본 연구에서는 이 중복성을 효과적으로 제거하여 이용하여 원 신경망의 성능과 원하는 범위안의 차이를 보이면서, 네트워크 연결의 수를 줄이는 방법을 제안하고 실험하였다. 특히, 재학습에 의하여 성능을 향상시키고, 각 계층별 차이를 고려하기 위하여 계층별 오류율을 할당하여 원하는 성능을 보장할 수 있는 간단한 방법을 제안하였다. 대표적인 영상인식 신경망구조인 FCN (전연결) 구조와 CNN (컨벌루션 신경망) 구조에서 대하여 실험한 결과 약 1/10 정도의 연결만으로도 원 신경망과 유사한 성능을 보일 수 있음을 확인하였다.

• 천문학회보
• /
• 제45권1호
• /
• pp.49.3-49.3
• /
• 2020

I will give an overview of the recent work in deriving cosmological constraints from deep learning methods applied to the large scale distribution of galaxies. I will specifically highlight the success of convolutional neural networks in linking the morphology of the large scale matter distribution to dark energy parameters and modified gravity scenarios.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2018년도 춘계학술발표대회
• /
• pp.377-380
• /
• 2018

인간의 추론 능력이란 문제에 주어진 조건을 보고 문제 해결에 필요한 것이 무엇인지를 논리적으로 생각해 보는 것으로 문제 상황 속에서 일정한 규칙이나 성질을 발견하고 이를 수학적인 방법으로 법칙을 찾아내거나 해결하는 능력을 말한다. 이러한 인간인지 능력과 유사한 인공지능 시스템을 개발하는데 있어서 핵심적 도전은 비구조적 데이터(unstructured data)로부터 그 개체들(object)과 그들간의 관계(relation)에 대해 추론하는 능력을 부여하는 것이라고 할 수 있다. 지금까지 딥러닝(deep learning) 방법은 구조화 되지 않은 데이터로부터 문제를 해결하는 엄청난 진보를 가져왔지만, 명시적으로 개체간의 관계를 고려하지 않고 이를 수행해왔다. 최근 발표된 구조화되지 않은 데이터로부터 복잡한 관계 추론을 수행하는 심층신경망(deep neural networks)은 관계추론(relational reasoning)의 시도를 이해하는데 기대할 만한 접근법을 보여주고 있다. 그 첫 번째는 관계추론을 위한 간단한 신경망 모듈(A simple neural network module for relational reasoning) 인 RN(Relation Networks)이고, 두 번째는 시각적 관찰을 기반으로 실제대상의 미래 상태를 예측하는 범용 목적의 VIN(Visual Interaction Networks)이다. 관계 추론을 수행하는 이들 심층신경망(deep neural networks)은 세상을 객체(objects)와 그들의 관계(their relations)라는 체계로 분해하고, 신경망(neural networks)이 피상적으로는 매우 달라 보이지만 근본적으로는 공통관계를 갖는 장면들에 대하여 객체와 관계라는 새로운 결합(combinations)을 일반화할 수 있는 강력한 추론 능력(powerful ability to reason)을 보유할 수 있다는 것을 보여주고 있다. 본 논문에서는 관계 추론을 수행하는 심층신경망(deep neural networks) 중에서 Sort-of-CLEVR 데이터 셋(dataset)을 사용하여 RN(Relation Networks)의 성능을 재현 및 관찰해 보았으며, 더 나아가 파라미터(parameters) 튜닝을 통하여 RN(Relation Networks) 모델의 성능 개선방법을 제시하여 보았다.

• 전기전자학회논문지
• /
• 제24권4호
• /
• pp.969-974
• /
• 2020

Research into hyper parameter optimization (HPO) has recently revived with interest in models containing many hyper parameters, such as deep neural networks. In this paper, we introduce the most widely used HPO methods, such as grid search, random search, and Bayesian optimization, and investigate their characteristics through experiments. The MNIST data set is used to compare results in experiments to find the best method that can be used to achieve higher accuracy in a relatively short time simulation. The learning rate and weight decay have been chosen for this experiment because these are the commonly used parameters in this kind of experiment.

• 대한임베디드공학회논문지
• /
• 제14권6호
• /
• pp.287-294
• /
• 2019

With the automation of production lines in the manufacturing industry, the importance of real-time fault diagnosis of facility is increasing. In this paper, we propose a fault diagnosis algorithm of LM (Linear Motion)-guide based on deep learning using vibration signals. Generally, in order to guarantee the performance of the deep learning, it is necessary to have a sufficient amount of data, but in a manufacturing industry, it is often difficult to obtain enough data due to physical and time constraints. To solve this problem, we propose a convolutional neural networks (CNN) model based on transfer learning. In addition, the spectrogram image is input to the CNN to reflect the frequency characteristic of the vibration signals with time. The performance of fault diagnosis according to various load condition and transfer learning method was compared and evaluated by experiments. The results showed that the proposed algorithm exhibited an excellent performance.

• 한국멀티미디어학회논문지
• /
• 제25권11호
• /
• pp.1653-1671
• /
• 2022

Crop diseases affect crop production, more than 30 billion USD globally. We proposed a classification study of crop species and diseases using deep learning algorithms for corn, cucumber, pepper, and strawberry. Our study has three steps of species classification, disease detection, and disease classification, which is noteworthy for using captured images without additional processes. We designed deep learning approach of deep learning convolutional neural networks based on Mask R-CNN model to classify crop species. Inception and Resnet models were presented for disease detection and classification sequentially. For classification, we trained Mask R-CNN network and achieved loss value of 0.72 for crop species classification and segmentation. For disease detection, InceptionV3 and ResNet101-V2 models were trained for nodes of crop species on 1,500 images of normal and diseased labels, resulting in the accuracies of 0.984, 0.969, 0.956, and 0.962 for corn, cucumber, pepper, and strawberry by InceptionV3 model with higher accuracy and AUC. For disease classification, InceptionV3 and ResNet 101-V2 models were trained for nodes of crop species on 1,500 images of diseased label, resulting in the accuracies of 0.995 and 0.992 for corn and cucumber by ResNet101 with higher accuracy and AUC whereas 0.940 and 0.988 for pepper and strawberry by Inception.

• 한국도로학회논문집
• /
• 제19권4호
• /
• pp.1-7
• /
• 2017

PURPOSES : This study aims to improve complex modeling of multivariable, nonlinear, and overdispersion data with an artificial neural network that has been a problem in the civil and transport sectors. METHODS: Deep learning, which is a technique employing artificial neural networks, was applied for developing a large bus fuel consumption model as a case study. Estimation characteristics and accuracy were compared with the results of conventional multiple regression modeling. RESULTS : The deep learning model remarkably improved estimation accuracy of regression modeling, from R-sq. 18.76% to 72.22%. In addition, it was very flexible in reflecting large variance and complex relationships between dependent and independent variables. CONCLUSIONS : Deep learning could be a new alternative that solves general problems inherent in conventional statistical methods and it is highly promising in planning and optimizing issues in the civil and transport sectors. Extended applications to other fields, such as pavement management, structure safety, operation of intelligent transport systems, and traffic noise estimation are highly recommended.

• 로봇학회논문지
• /
• 제17권2호
• /
• pp.133-141
• /
• 2022

This paper introduces model compression algorithms which make a deep neural network smaller and faster for embedded systems. The model compression algorithms can be largely categorized into pruning, quantization and knowledge distillation. In this study, gradual pruning, quantization aware training, and knowledge distillation which learns the activation boundary in the hidden layer of the teacher neural network are integrated. As a large deep neural network is compressed and accelerated by these algorithms, embedded computing boards can run the deep neural network much faster with less memory usage while preserving the reasonable accuracy. To evaluate the performance of the compressed neural networks, we evaluate the size, latency and accuracy of the deep neural network, DenseNet201, for image classification with CIFAR-10 dataset on the NVIDIA Jetson Xavier.

• 스마트미디어저널
• /
• 제9권1호
• /
• pp.45-50
• /
• 2020

딥 러닝의 발전에 따라 추천시스템에서 딥 러닝 기반의 인공신경망을 활용한 연구가 활발히 진행되고 있다. 특히, RNN(Recurrent Neural Network)기반의 추천시스템은 데이터의 순차적 특성을 고려하기 때문에 추천시스템에서 좋은 성과를 보여주고 있다. 본 연구는 RNN기반의 알고리즘인 GRU(Gated Recurrent Unit)와 세션 기반 병렬 미니배치(Session Parallel mini-batch)기법을 활용한 여행경로 추천 시스템을 제안한다. 본 연구는 top1과 bpr(Bayesian personalized ranking) 오차함수의 앙상블을 통해 추천 성과를 향상시켰다. 또한, 데이터 내에 순차적인 특성을 고려한 RNN기반 추천 시스템은 여행경로에 내재된 여행지의 의미가 반영된 추천이 이루어진다는 것을 확인되었다.

• 반도체디스플레이기술학회지
• /
• 제19권3호
• /
• pp.124-129
• /
• 2020

Autonomous underwater vehicle makes attracts to many researchers. This paper proposes a convolutional neural network (CNN) based fish detection method. Since there are not enough data sets in the process of training, overfitting problem can be occurred in deep learning. To solve the problem, we apply the dropout algorithm to simplify the model. Experimental result showed that the implemented method is promising, and the effectiveness of identification by dropout approach is highly enhanced.