• 한국의학물리학회지:의학물리
• /
• 제30권2호
• /
• pp.39-48
• /
• 2019

Deep learning has been applied to various medical data. In particular, current deep learning models exhibit remarkable performance at specific tasks, sometimes offering higher accuracy than that of experts for discriminating specific diseases from medical images. The current status of deep learning applications to molecular imaging can be divided into a few subtypes in terms of their purposes: differential diagnostic classification, enhancement of image acquisition, and image-based quantification. As functional and pathophysiologic information is key to molecular imaging, this review will emphasize the need for accurate biomarker acquisition by deep learning in molecular imaging. Furthermore, this review addresses practical issues that include clinical validation, data distribution, labeling issues, and harmonization to achieve clinically feasible deep learning models. Eventually, deep learning will enhance the role of theranostics, which aims at precision targeting of pathophysiology by maximizing molecular imaging functional information.

• International Journal of Advanced Culture Technology
• /
• 제7권4호
• /
• pp.295-301
• /
• 2019

Since the third artificial intelligence boom was triggered by deep learning, it has been 10 years. It is time to analyze and discuss the research trends of deep learning for the stable development of AI. In this regard, this study systematically analyzes the trends of research on deep learning over the past 10 years. We collected research literature on deep learning and performed LDA based topic modeling analysis. We analyzed trends by topic over 10 years. We have also identified differences among the major research countries, China, the United States, South Korea, and United Kingdom. The results of this study will provide insights into research direction on deep learning in the future, and provide implications for the stable development strategy of deep learning.

• International journal of advanced smart convergence
• /
• 제12권4호
• /
• pp.164-170
• /
• 2023

With the recent advances in AI (artificial intelligence) and HPC (high-performance computing) technologies, deep learning is proliferated in various domains of the 4th industrial revolution. As the workload volume of deep learning increasingly grows, analyzing the memory reference characteristics becomes important. In this article, we analyze the memory reference traces of deep learning workloads in comparison with traditional workloads specially focusing on read and write operations. Based on our analysis, we observe some unique characteristics of deep learning memory references that are quite different from traditional workloads. First, when comparing instruction and data references, instruction reference accounts for a little portion in deep learning workloads. Second, when comparing read and write, write reference accounts for a majority of memory references, which is also different from traditional workloads. Third, although write references are dominant, it exhibits low reference skewness compared to traditional workloads. Specifically, the skew factor of write references is small compared to traditional workloads. We expect that the analysis performed in this article will be helpful in efficiently designing memory management systems for deep learning workloads.

• 의학교육논단
• /
• 제11권1호
• /
• pp.35-43
• /
• 2009

Learning strategies are defined as behaviors and thoughts that a learner engages in during learning and that are intended to influence the learner's encoding process. Today, demands for teaching how to learn increase, because there is a lot of complex material which is delivered to students. But learning strategies shouldn't be identified as tricks of students for achieving high scores in exams. Cognitive researchers and theorists assume that learning strategies are related to two types of learning processing, which are described as 'surface learning' and 'deep learning'. In addition learning strategies are associated with learning motivation. Students with 'meaning orientation' who struggle for deep learning, are intrinsically motivated, whereas students with 'reproduction orientation' or 'achieving orientation' are extrinsically motivated. Therefore, to foster active learning and intrinsic motivation of students, it isn't enough to just teach how to learn. Changes of curriculum and assessment methods, that stimulate deep learning and curiosity of students are needed with educators and learners working cooperatively.

• Journal of Information Processing Systems
• /
• 제16권5호
• /
• pp.1001-1007
• /
• 2020

In recent years, machine learning algorithms are continuously being used and expanded in various fields, such as facial recognition, signal processing, personal authentication, and stock prediction. In particular, various algorithms, such as deep learning, reinforcement learning, and Q-learning, are continuously being improved. Among these algorithms, the expansion of deep learning is rapidly changing. Nevertheless, machine learning algorithms have not yet been applied in several fields, such as personal authentication technology. This technology is an essential tool in the digital information era, walking recognition technology as promising biometrics, and technology for solving state-space problems. Therefore, algorithm technologies of deep learning, reinforcement learning, and Q-learning, which are typical machine learning algorithms in various fields, such as agricultural technology, personal authentication, wireless network, game, biometric recognition, and image recognition, are being improved and expanded in this paper.

• 지능정보연구
• /
• 제26권4호
• /
• pp.27-65
• /
• 2020

많은 정보통신기술 기업들은 자체적으로 개발한 인공지능 기술을 오픈소스로 공개하였다. 예를 들어, 구글의 TensorFlow, 페이스북의 PyTorch, 마이크로소프트의 CNTK 등 여러 기업들은 자신들의 인공지능 기술들을 공개하고 있다. 이처럼 대중에게 딥러닝 오픈소스 소프트웨어를 공개함으로써 개발자 커뮤니티와의 관계와 인공지능 생태계를 강화하고, 사용자들의 실험, 적용, 개선을 얻을 수 있다. 이에 따라 머신러닝 분야는 급속히 성장하고 있고, 개발자들 또한 여러가지 학습 알고리즘을 재생산하여 각 영역에 활용하고 있다. 하지만 오픈소스 소프트웨어에 대한 다양한 분석들이 이루어진 데 반해, 실제 산업현장에서 딥러닝 오픈소스 소프트웨어를 개발하거나 활용하는데 유용한 연구 결과는 미흡한 실정이다. 따라서 본 연구에서는 딥러닝 프레임워크 사례연구를 통해 해당 프레임워크의 도입 전략을 도출하고자 한다. 기술-조직-환경 프레임워크를 기반으로 기존의 오픈 소스 소프트웨어 도입과 관련된 연구들을 리뷰하고, 이를 바탕으로 두 기업의 성공 사례와 한 기업의 실패 사례를 포함한 총 3 가지 기업의 도입 사례 분석을 통해 딥러닝 프레임워크 도입을 위한 중요한 5가지 성공 요인을 도출하였다: 팀 내 개발자의 지식과 전문성, 하드웨어(GPU) 환경, 데이터 전사 협력 체계, 딥러닝 프레임워크 플랫폼, 딥러닝 프레임워크 도구 서비스. 그리고 도출한 성공 요인을 실현하기 위한 딥러닝 프레임워크의 단계적 도입 전략을 제안하였다: 프로젝트 문제 정의, 딥러닝 방법론이 적합한 기법인지 확인, 딥러닝 프레임워크가 적합한 도구인지 확인, 기업의 딥러닝 프레임워크 사용, 기업의 딥러닝 프레임워크 확산. 본 연구를 통해 각 산업과 사업의 니즈에 따라, 딥러닝 프레임워크를 개발하거나 활용하고자 하는 기업에게 전략적인 시사점을 제공할 수 있을 것이라 기대된다.

• 반도체디스플레이기술학회지
• /
• 제19권2호
• /
• pp.60-67
• /
• 2020

Recently, artificial intelligence related technologies including machine learning are being applied to various fields, and the demand is also increasing. In particular, with the development of AR, VR, and MR technologies related to image processing, the utilization of computer vision based on deep learning has increased. The algorithms for object recognition and detection based on deep learning required for image processing are diversified and advanced. Accordingly, problems that were difficult to solve with the existing methodology were solved more simply and easily by using deep learning. This paper introduces various deep learning-based object recognition and extraction algorithms used to detect and recognize various objects in an image and analyzes the technologies that attract attention.

• ETRI Journal
• /
• 제42권2호
• /
• pp.292-304
• /
• 2020

As side-channel analysis and machine learning algorithms share the same objective of classifying data, numerous studies have been proposed for adapting machine learning to side-channel analysis. However, a drawback of machine learning algorithms is that their performance depends on human engineering. Therefore, recent studies in the field focus on exploiting deep learning algorithms, which can extract features automatically from data. In this study, we survey recent advances in deep learning-based side-channel analysis. In particular, we outline how deep learning is applied to side-channel analysis, based on deep learning architectures and application methods. Furthermore, we describe its properties when using different architectures and application methods. Finally, we discuss our perspective on future research directions in this field.

• 한국안전학회지
• /
• 제34권6호
• /
• pp.50-57
• /
• 2019

In this study, various types of deep learning models that have been proposed recently are classified according to data input / output types and analyzed to find the deep learning model suitable for constructing a crack detection model. First the deep learning models are classified into image classification model, object segmentation model, object detection model, and instance segmentation model. ResNet-101, DeepLab V2, Faster R-CNN, and Mask R-CNN were selected as representative deep learning model of each type. For the comparison, ResNet-101 was implemented for all the types of deep learning model as a backbone network which serves as a main feature extractor. The four types of deep learning models were trained with 500 crack images taken from real concrete structures and collected from the Internet. The four types of deep learning models showed high accuracy above 94% during the training. Comparative evaluation was conducted using 40 images taken from real concrete structures. The performance of each type of deep learning model was measured using precision and recall. In the experimental result, Mask R-CNN, an instance segmentation deep learning model showed the highest precision and recall on crack detection. Qualitative analysis also shows that Mask R-CNN could detect crack shapes most similarly to the real crack shapes.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제14권11호
• /
• pp.4246-4267
• /
• 2020

As the network goes deep into all aspects of people's lives, the number and the complexity of network traffic is increasing, and traffic classification becomes more and more important. How to classify them effectively is an important prerequisite for network management and planning, and ensuring network security. With the continuous development of deep learning, more and more traffic classification begins to use it as the main method, which achieves better results than traditional classification methods. In this paper, we provide a comprehensive review of network traffic classification based on deep learning. Firstly, we introduce the research background and progress of network traffic classification. Then, we summarize and compare traffic classification based on deep learning such as stack autoencoder, one-dimensional convolution neural network, two-dimensional convolution neural network, three-dimensional convolution neural network, long short-term memory network and Deep Belief Networks. In addition, we compare traffic classification based on deep learning with other methods such as based on port number, deep packets detection and machine learning. Finally, the future research directions of network traffic classification based on deep learning are prospected.