• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.10 no.4
• /
• pp.257-267
• /
• 2017

Compressed Sensing(CS) deals with linear inverse problems. The theoretical results of CS have had an impact on inference problems and presented amazing research achievements in the related fields including signal processing and information theory. However, in order for CS to be applied in practical environments, there are two significant challenges to be solved. One is to guarantee in real time recovery of CS signals, and the other is that the signals have to be sparse. To this end, the latest researches using deep learning technology have emerged. In this paper, we consider CS problems based on deep learning and discuss the latest research results. And the approaches for CS signal reconstruction using deep learning show superior results in terms of recovery time and performance. It is expected that the approaches for CS reconstruction using deep learning shown in recent studies can not only raise the possibility of utilization of CS, but also be highly exploited in the fields of signal processing and communication areas.

• International Journal of Advanced Culture Technology
• /
• v.7 no.4
• /
• pp.56-62
• /
• 2019

Recently, research using artificial neural networks has further expanded the field of neural network optimization and automatic structuring from improving inference accuracy. The performance of the machine learning algorithm depends on how the hyperparameters are configured. Open-source hyperparameter optimization software can be an important step forward in improving the performance of machine learning algorithms. In this paper, we review open-source hyperparameter optimization softwares.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2020.11a
• /
• pp.934-936
• /
• 2020

기계학습 모델의 전체 구조를 쉽게 파악하고 추론할 수 있으며, 추론 과정 중에 멈춰서 중간결과를 확인할 수 있는 디버깅, 그리고 customizing 까지 지원하여 기계학습에 더 익숙해지고 더 나아가, 실제로 활용해보는 GUI Platform 구현

• Journal of Korea Multimedia Society
• /
• v.22 no.5
• /
• pp.547-557
• /
• 2019

To extract the depth map from a single image, a number of CNN-based deep learning methods have been performed in recent research. In this study, the GAN structure of Pix2Pix is maintained. this model allows to converge well, because it has the structure of the generator and the discriminator. But the convolution in this model takes a long time to compute. So we change the convolution form in the generator to a depthwise convolution to improve the speed while preserving the result. Thus, the seven down-sizing convolutional hidden layers in the generator U-Net are changed to depthwise convolution. This type of convolution decreases the number of parameters, and also speeds up computation time. The proposed model shows similar depth map prediction results as in the case of the existing structure, and the computation time in case of a inference is decreased by 64%.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.16 no.1
• /
• pp.1-8
• /
• 2021

Detecting Out-of-Distribution (OOD) data is fundamentally required when Deep Neural Network (DNN) is applied to real-world AI such as autonomous driving. However, modern DNNs are quite vulnerable to the over-confidence problem even if the test data are far away from the trained data distribution. To solve the problem, this paper proposes a novel Adversarial-Mixup training method to let the DNN model be more robust by detecting OOD data effectively. Experimental results show that the proposed Adversarial-Mixup method improves the overall performance of OOD detection by 78% comparing with the State-of-the-Art methods. Furthermore, we show that the proposed method can alleviate the over-confidence problem by reducing the confidence score of OOD data than the previous methods, resulting in more reliable and robust DNNs.

• Journal of Korea Multimedia Society
• /
• v.25 no.5
• /
• pp.676-684
• /
• 2022

As realistic media are widespread in various image processing areas, image or video compression is one of the key technologies to enable real-time applications with limited network bandwidth. Generally, image or video compression cause the unnecessary compression artifacts, such as blocking artifacts and ringing effects. In this study, we propose a Deep Residual Channel-attention Network, so called DRCAN, which consists of an input layer, a feature extractor and an output layer. Experimental results showed that the proposed DRCAN can reduced the total memory size and the inference time by as low as 47% and 59%, respectively. In addition, DRCAN can achieve a better peak signal-to-noise ratio and structural similarity index measure for compressed images compared to the previous methods.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2022.11a
• /
• pp.234-236
• /
• 2022

딥러닝 원격 컴퓨팅 서비스(Deep Learning as a Service, DLaaS)가 널리 활용되면서 클라우드에서의 개인 정보 보호에 대한 우려가 커졌다. 신뢰실행환경(Trusted Execution Environment, TEE)는 운영체제의 접근까지 차단한 메인 프로세서의 보안 영역으로 DLaaS 환경에서의 개인 정보 보호 기법으로 채택되고 있다. 사용자의 데이터를 보호하면서 고성능 클라우드 환경을 활용하기 위해 신뢰실행환경을 활용한 딥러닝 모델 추론 연구들을 살펴보고자 한다.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.1
• /
• pp.392-412
• /
• 2018

Due to the semantic gap problem across different modalities, automatically retrieval from multimedia information still faces a main challenge. It is desirable to provide an effective joint model to bridge the gap and organize the relationships between them. In this work, we develop a deep image annotation and classification by fusing multi-modal semantic topics (DAC_mmst) model, which has the capacity for finding visual and non-visual topics by jointly modeling the image and loosely related text for deep image annotation while simultaneously learning and predicting the class label. More specifically, DAC_mmst depends on a non-parametric Bayesian model for estimating the best number of visual topics that can perfectly explain the image. To evaluate the effectiveness of our proposed algorithm, we collect a real-world dataset to conduct various experiments. The experimental results show our proposed DAC_mmst performs favorably in perplexity, image annotation and classification accuracy, comparing to several state-of-the-art methods.

• Journal of Ocean Engineering and Technology
• /
• v.20 no.3 s.70
• /
• pp.54-60
• /
• 2006

For optimal design of a deep-sea ocean mining collector system, based on self-propelled mining vehicle, it is imperative to develop and validate the dynamic model of a tracked vehicle traveling on soft deep seabed. The purpose of this paper is to evaluate the fidelity of the dynamic simulation model by means of response surface methodology. Various statistical techniques related to response surface methodology, such as outlier analysis, detection of interaction effect, analysis of variance, inference of the significance of design variables, and global sensitivity analysis, are examined. To obtain a plausible response surface model, maximum entropy sampling is adopted. From statistical analysis and prediction for dynamic responses of the tracked vehicle, conclusions will be drawn about the accuracy of the dynamic model and the performance of the response surface model.

• Nuclear Engineering and Technology
• /
• v.53 no.12
• /
• pp.4014-4021
• /
• 2021

If safety injection systems (SISs) do not work in the event of a loss-of-coolant accident (LOCA), the accident can progress to a severe accident in which the reactor core is exposed and the reactor vessel fails. Therefore, it is considered that a technology that provides recoverable maximum time for SIS actuation is necessary to prevent this progression. In this study, the corresponding time was defined as the golden time. To achieve the objective of accurately predicting the golden time, the prediction was performed using the deep fuzzy neural network (DFNN) with rule-dropout. The DFNN with rule-dropout has an architecture in which many of the fuzzy neural networks (FNNs) are connected and is a method in which the fuzzy rule numbers, which are directly related to the number of nodes in the FNN that affect inference performance, are properly adjusted by a genetic algorithm. The golden time prediction performance of the DFNN model with rule-dropout was better than that of the support vector regression model. By using the prediction result through the proposed DFNN with rule-dropout, it is expected to prevent the aggravation of the accidents by providing the maximum remaining time for SIS recovery, which failed in the LOCA situation.