• The Journal of the Korea Contents Association
• /
• v.10 no.8
• /
• pp.1-10
• /
• 2010

For virtual education, the multimodal learning environment with haptic feedback, termed 'immersive textbook', is necessary to enhance the learning effectiveness. However, the learning contents for immersive textbook are not widely available due to the constraints in creation and playback environments. To address this problem, we propose a framework for producing and displaying the multimodal contents for immersive textbook. Our framework provides an XML-based meta-language to produce the multimodal learning contents in the form of intuitive script. Thus it can help the user, without any prior knowledge of multimodal interactions, produce his or her own learning contents. The contents are then interpreted by script engine and delivered to the user by visual and haptic rendering loops. Also we implemented a prototype based on the aforementioned proposals and performed user evaluation to verify the validity of our framework.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.71-79
• /
• 2023

In this paper, we analyze the characteristics of machine learning workloads and, based on them, propose a distributed in-memory caching technique to improve the performance of machine learning workloads. The core of machine learning workload is model training, and model training is a computationally intensive task. Performing machine learning workloads in a Kubernetes-based cloud environment in which the computing framework and storage are separated can effectively allocate resources, but delays can occur because IO must be performed through network communication. In this paper, we propose a distributed in-memory caching technique to improve the performance of machine learning workloads performed in such an environment. In particular, we propose a new method of precaching data required for machine learning workloads into the distributed in-memory cache by considering Kubflow pipelines, a Kubernetes-based machine learning pipeline management tool.

• Journal of the Institute of Electronics and Information Engineers
• /
• v.54 no.5
• /
• pp.35-41
• /
• 2017

In this paper, we analyze the performance of the recently introduced Hint-knowledge distillation (KD) training approach based on the teacher-student framework for knowledge distillation and knowledge transfer. As a deep neural network (DNN) considered in this paper, the deep residual network (ResNet), which is currently regarded as the latest DNN, is used for the teacher-student framework. Therefore, when implementing the Hint-KD training, we investigate the impact on the weight of KD information based on the soften factor in terms of classification accuracy using the widely used open deep learning frameworks, Caffe. As a results, it can be seen that the recognition accuracy of the student model is improved when the fixed value of the KD information is maintained rather than the gradual decrease of the KD information during training.

• The Journal of the Convergence on Culture Technology
• /
• v.5 no.2
• /
• pp.375-380
• /
• 2019

In this paper, we study learning process mining and analytic technology based on open education platform. A study on mining through personal learning history log data based on an open education platform such as MOOC which is growing in interest recently. This technology is to design and implement a learning process mining framework for discovering and analyzing meaningful learning processes and knowledge from learning history log data. Learning process mining framework technology is a technique for expressing, extracting, analyzing and visualizing the learning process to provide learners with improved learning processes and educational services.

• Broadcasting and Media Magazine
• /
• v.28 no.1
• /
• pp.13-28
• /
• 2023

이미지 압축은 이미지 및 영상처리에서 주요한 역할을 하며, 자율주행, 클라우드, 영상 송출 등의 분야에서 빅데이터를 처리해야 하는 수요가 늘어남에 따라 지속적인 연구가 진행 중이다. 그 중심에는 딥러닝(deep learning)의 발전이 자리잡고 있으며, 심층 신경망(deep neural network)을 효과적으로 학습하는 알고리즘들을 적용한 논문들은 기존 압축 포맷인 JPEG, JPEG 2000, MPEG 등의 압축 성능을 뛰어넘는 결과를 보여 주고 있다. 이에 따라 JPEG AI는 딥러닝 기반 학습 이미지 압축의 표준을 제정하는 일을 진행 중이다. 본 기고에서는 JPEG AI가 표준화하고자 하는 기술과 JPEG AI에 제안한 압축 프레임워크들을 분석하고, 활용 사례들을 소개하여 JPEG AI 기반 학습 이미지 압축 모델의 동향에 대해 알아보고자 한다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.311-313
• /
• 2023

딥러닝에 대한 관심이 증가함에 따라 다양한 분야의 연구자 사이에 딥러닝 모델의 적용 및 재현이 중요한 작업으로 자리잡았다. 하지만 모델을 재현하고 활용하는데 있어 다양한 환경과 자원의 한계가 발생하여 문제가 되고 있다. 이러한 문제를 해결하기 위해 본 논문에서는 국가연구데이터커먼즈체계인 KRDC 프레임워크를 활용하여 딥러닝 학습 모델의 재현 방안을 제안하였다. 이를 통해 딥러닝 연구에 익숙하지 않은 사용자도 학습 모델의 적용 및 활용을 용이하게 할 수 있음을 확인하였다. KRDC 프레임워크는 사용자가 원하는 데이터와 태스크를 정의하고, 워크플로우로 구성, 학습 모델의 재현 및 활용을 지원한다.

• Journal of Software Assessment and Valuation
• /
• v.17 no.1
• /
• pp.101-113
• /
• 2021

In this paper, we introduce a speaker overlap system and look at the process of converting the existed system on the specific framework of artificial intelligence. Speaker overlap is when two or more speakers speak at the same time during a conversation, and can lead to performance degradation in the fields of speech recognition or speaker recognition, and a lot of research is being conducted because it can prevent performance degradation. Recently, as application of artificial intelligence is increasing, there is a demand for switching between artificial intelligence frameworks. However, when switching frameworks, performance degradation is observed due to the unique characteristics of each framework, making it difficult to switch frameworks. In this paper, the process of converting the speaker overlap detection system based on the Keras framework to the pytorch-based system is explained and considers components. As a result of the framework switching, the pytorch-based system showed better performance than the existing Keras-based speaker overlap detection system, so it can be said that it is valuable as a fundamental study on systematic framework conversion.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.25-28
• /
• 2014

에너지 자립섬은 외부 전력의 유입이 어려운 상황에서 풍력/태양광 발전과 같은 재생 에너지를 주요 발전원으로 운영하는 섬이다. 에너지 자립섬의 운영을 위해서는 전력 수요와 공급량을 예측하여 발전기, 송배전 시스템, ESS 등의 운영 계획 수립이 필요한데 수요 및 공급의 예측은 기상 상황 및 시간 등의 다양한 요소에 영향을 받으므로 예측이 어렵다. 이러한 특성을 감안하여 효율적인 전력망 운영을 위해 기계 학습을 기반으로 한 스마트 그리드 운영 프레임워크의 활용을 통해 이 문제를 해결하고자 한다. 본 논문에서는 자립섬 운영 계획 수립에 필요한 구성 요소를 파악하고 요소들 간의 연계 관계를 분석하여 운영 시스템의 프레임워크 설계안을 제시한다.

• Journal of The Korean Association of Information Education
• /
• v.26 no.2
• /
• pp.129-140
• /
• 2022

Our society is undergoing rapid changes due to COVID-19, and in particular, online learning using digital technology is being tried in various forms in the educational field. A change has occurred. However, the limitations of distance learning, such as reduced learning immersion in non-face-to-face educational situations, lack of interaction between teachers and learners, and lower basic academic ability, are constantly being raised, and an appropriate educational strategy is needed to solve these problems. This study focused on the concept of 'Metaverse' based on the interaction between the virtual world and the real world, and tried to verify the effectiveness of educational activities based on it. In detail, we propose an educational framework for realizing flipped learning in the Metaverse Virtual Classroom, and a frame developed by measuring the learning immersion of a single group with a teaching/learning program developed based on this. The effectiveness of the work was verified. When the metaverse platform-based flip learning framework and education program proposed in this study were applied, it was confirmed that learners' immersion in learning was improved.

• The KIPS Transactions:PartA
• /
• v.17A no.5
• /
• pp.249-258
• /
• 2010

The utilization of robot application is growing up in recent years, but there is a constraint to execute various application on the robot because of difference of robot resource. This paper presents the framework in order to solve the resource constraint by sharing resources with other devices near by robot. The framework defines common factors that are needed to collaboration work and provides APIs in order to implement robot application easily. Furthermore, We show the working flow of framework with physical training application using robot by example. The application shows how to collaborated work between robot and other devices through network.