• Journal of Broadcast Engineering
• /
• v.27 no.4
• /
• pp.511-518
• /
• 2022

Supervised learning based on deep learning has made a leap forward in various application fields. However, many supervised learning methods work under the common assumption that training and test data are extracted from the same distribution. If it deviates from this constraint, the deep learning network trained in the training domain is highly likely to deteriorate rapidly in the test domain due to the distribution difference between domains. Domain adaptation is a methodology of transfer learning that trains a deep learning network to make successful inferences in a label-poor test domain (i.e., target domain) based on learned knowledge of a labeled-rich training domain (i.e., source domain). In particular, the unsupervised domain adaptation technique deals with the domain adaptation problem by assuming that only image data without labels in the target domain can be accessed. In this paper, we explore the unsupervised domain adaptation techniques.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.10
• /
• pp.55-65
• /
• 2023

In this paper, we apply KeyBERT(Keyword extraction with Bidirectional Encoder Representations of Transformers) algorithm-driven topic extraction and topic frequency analysis to deep learning and reinforcement learning research to discover the rapidly changing trends in them. First, we crawled abstracts of research papers on deep learning and reinforcement learning, and temporally divided them into two groups. After pre-processing the crawled data, we extracted topics using KeyBERT algorithm, and then analyzed the extracted topics in terms of topic occurrence frequency. This analysis reveals that there are distinct trends in research work of all analyzed algorithms and applications, and we can clearly tell which topics are gaining more interest. The analysis also proves the effectiveness of the utilized topic extraction and topic frequency analysis in research trend analysis, and this trend analysis scheme is expected to be used for research trend analysis in other research fields. In addition, the analysis can provide insight into how deep learning will evolve in the near future, and provide guidance for select research topics and methodologies by informing researchers of research topics and methodologies which are recently attracting attention.

• Journal of Internet of Things and Convergence
• /
• v.9 no.6
• /
• pp.119-124
• /
• 2023

Deep learning has recently demonstrated conspicuous efficacy across diverse domains than traditional machine learning techniques, as the most popular approach for pattern recognition. The classification problems for tabular data, however, are remain for the area of traditional machine learning. This paper introduces a novel network module designed to tabular data into high-dimensional tensors. The module is integrated into conventional deep learning networks and subsequently applied to the classification of structured data. The proposed method undergoes training and validation on four datasets, culminating in an average accuracy of 90.22%. Notably, this performance surpasses that of the contemporary deep learning model, TabNet, by 2.55%p. The proposed approach acquires significance by virtue of its capacity to harness diverse network architectures, renowned for their superior performance in the domain of computer vision, for the analysis of tabular data.

• Journal of Advanced Research in Ocean Engineering
• /
• v.3 no.3
• /
• pp.136-148
• /
• 2017

The paper presents dual arm ROV manipulation using deep reinforcement learning. The purpose of this underwater manipulator is to investigate and excavate natural resources in ocean, finding lost aircraft blackboxes and for performing other extremely dangerous tasks without endangering humans. This research work emphasizes on a self-learning approach using Deep Reinforcement Learning (DRL). DRL technique allows ROV to learn the policy of performing manipulation task directly, from raw image data. Our proposed architecture maps the visual inputs (images) to control actions (output) and get reward after each action, which allows an agent to learn manipulation skill through trial and error method. We have trained our network in simulation. The raw images and rewards are directly provided by our simple Lua simulator. Our simulator achieve accuracy by considering underwater dynamic environmental conditions. Major goal of this research is to provide a smart self-learning way to achieve manipulation in highly dynamic underwater environment. The results showed that a dual robotic arm trained for a 3DOF movement successfully achieved target reaching task in a 2D space by considering real environmental factor.

• Journal of Korea Multimedia Society
• /
• v.21 no.8
• /
• pp.864-875
• /
• 2018

In this paper, we propose a measurement method for height of white light scanning interferometer using deep learning. In order to measure the fine surface shape, a three-dimensional surface shape measurement technique is required. A typical example is a white light scanning interferometer. In order to calculate the surface shape from the measurement image of the white light scanning interferometer, the height of each pixel must be calculated. In this paper, we propose a neural network for height calculation and use virtual data generation method to train this neural network. The accuracy was measured by inputting 57 actual data to the neural network which had completed the learning. We propose two new functions for accuracy measurement. We have analyzed the cases where there are many errors among the accuracy calculation values, and it is confirmed that there are many errors when there is no interference fringe or outside the learned range. We confirmed that the proposed neural network works correctly in most cases. We expect better results if we improve the way we generate learning data.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_2
• /
• pp.929-937
• /
• 2020

In this manuscript, the UNet++ model, which is one of the representative deep learning techniques for semantic segmentation, was used to detect changes in temporal satellite images. To analyze the learning results according to various loss functions, we evaluated the change detection results using trained UNet++ models by binary cross entropy and the Jaccard coefficient. In addition, the learning results of the deep learning model were analyzed compared to existing pixel-based change detection algorithms by using WorldView-3 images. In the experiment, it was confirmed that the performance of the deep learning model could be determined depending on the characteristics of the loss function, but it showed better results compared to the existing techniques.

• International journal of advanced smart convergence
• /
• v.9 no.2
• /
• pp.90-104
• /
• 2020

It is difficult for non-experts to build machine learning (ML) models at the level that satisfies their needs. Deep learning models are even more challenging because it is unclear how to improve the model, and a trial-and-error approach is not feasible since training these models are time-consuming. To assist these novice users, we examined how interactive and explainable feedback while training a deep learning network can contribute to model performance and users' satisfaction, focusing on the data preparation process. We conducted a user study with 31 participants without expertise, where they were asked to improve the accuracy of a deep learning model, varying feedback conditions. While no significant performance gain was observed, we identified potential barriers during the process and found that interactive and explainable feedback provide complementary benefits for improving users' understanding of ML. We conclude with implications for designing an interface for building ML models for novice users.

• Annual Conference of KIPS
• /
• 2016.10a
• /
• pp.615-616
• /
• 2016

최근 4차 산업으로 패러다임이 변화함에 따라 SW산업이 더욱 중요하게 되었다. 이에 따라 전 세계적으로 코딩 교육에 대한 수요도 증가하게 되었고 기업에서도 SW를 잘 만들기 위한 코드 관리 중요성도 증가하게 되었다. 많은 양의 프로그래밍 소스코드를 사람이 일일이 채점하고 관리하는 것은 사실상 불가능하기 때문에 이러한 문제를 해결할 수 있는 코드 평가 시스템이 요구되고 있다. 하지만 어떤 코드가 좋은 코드인지 코드를 어떻게 평가해야하는지에 대한 명확한 기준은 없으며 이에 대한 연구도 부족한 상황이다. 최근에 주목 받고 있는 Deep Learning 기술은 이미지 처리, 자연어 처리등 기존의 Machine Learning 알고리즘이 냈던 성과보다 훨씬 뛰어난 성과를 내고 있다. 하지만 Programming language 영역에서는 아직 깊이 연구된 바가 없다. 따라서 본 연구에서는 Deep Learning 기술로 알려진 Convolutional Neural Network의 변형된 형태엔 Tree-based Convolutional Neural Network를 사용하여 프로그래밍 소스코드를 분석, 분류하는 알고리즘 및 코드의 Representation Learning에 대한 연구를 진행함으로써 이러한 문제를 해결하고자 한다.

• International Journal of Computer Science & Network Security
• /
• v.21 no.8
• /
• pp.238-246
• /
• 2021

The rapid rise of the Internet and social media has resulted in a large number of text-based reviews being placed on sites such as social media. In the age of social media, utilizing machine learning technologies to analyze the emotional context of comments aids in the understanding of QoS for any product or service. The classification and analysis of user reviews aids in the improvement of QoS. (Quality of Services). Machine Learning algorithms have evolved into a powerful tool for analyzing user sentiment. Unlike traditional categorization models, which are based on a set of rules. In sentiment categorization, Bidirectional Long Short-Term Memory (BiLSTM) has shown significant results, and Convolution Neural Network (CNN) has shown promising results. Using convolutions and pooling layers, CNN can successfully extract local information. BiLSTM uses dual LSTM orientations to increase the amount of background knowledge available to deep learning models. The suggested hybrid model combines the benefits of these two deep learning-based algorithms. The data source for analysis and classification was user reviews of Indian Railway Services on Twitter. The suggested hybrid model uses the Keras Embedding technique as an input source. The suggested model takes in data and generates lower-dimensional characteristics that result in a categorization result. The suggested hybrid model's performance was compared using Keras and Word2Vec, and the proposed model showed a significant improvement in response with an accuracy of 95.19 percent.

• Journal of KIBIM
• /
• v.10 no.2
• /
• pp.21-28
• /
• 2020

As the number of aging bridges increases, more studies are being conducted on developing effective and reliable methods for the assessment and maintenance of bridges. With the advancement in new sensing systems and data learning techniques through AI technology, there is growing interests in how to evaluate bridges using these advanced techniques. This paper presents a CNN(Convolution Neural Network) deep learning based technique for evaluating the damage existence and for estimating the damage location in PSC bridges using static strain data. Simulation studies were conducted to investigate the proposed method with error analysis. Damage was simulated as the reduction in the stiffness of a finite element. A data learning model was constructed by applying the CNN technique as a type of deep learning. The damage status and its location were estimated using data set built through simulation. It was assumed that the strain gauges were installed in a regular interval under the PSC bridge girders. In order to increase the accuracy in evaluating damage, the squared error between the intact and measured strains are computed and applied for training the data model. Considering the damage occurring near the supports, the results of error analysis were compared according to whether strain data near the supports were included.