• 한국컴퓨터산업학회논문지
• /
• 제2권9호
• /
• pp.1139-1146
• /
• 2001

최근 웹 기술의 확산과 정보기술의 발달로 교육분야에서 컴퓨터의 가치는 더욱 높아지고 있으며, 기존의 교실위주의 수업보다 시간적 공간적 제약을 덜 받게 되는 WBI(Web Based Instruction)는 무한한 발전 가능성을 제시하고 있다. 하지만 기존의 WBI에서는 교수가 수업 평가 후 학생에게 수업의 결과를 알려주는 형식이었고, 또한 학습자에게는 일률적으로 평가가 적용되어 학생의 수준에 맞는 수업을 할 수가 없었으며, 교수의 입장에서도 학생들의 평가 결과가 수업의 질을 높이거나 내용을 변경시킬 적당한 근거 제시의 어려움을 가지고 있었다. 본 논문에서는 수준별 학습을 위한 단계별 예비 테스트와 학습 후 다양한 방법으로 테스트의 평가내용을 시각적으로 제시하고, 또한 교수가 평가의 준거를 입력하고 학습자의 평가결과와 교수의 평가준거를 비교할 수 있는 교수지원 컴포넌트를 설계 구현한다. 이는 체계적인 평가 방법론이 되고 학습의 패러다임을 바꾸거나 과목을 변경할 경우, 그 결과에 따라 교수 방법의 변화나 수업내용을 변경하고자할 때 용이하다. 또한 유사한 다른 패러다임의 WBI 시스템에서도 이미 개발된 컴포넌트를 사용함으로써, 유사 응용 시스템의 개발 시 제시된 컴퍼넌트 사용의 용이성과 이식성, 재사용성을 높일 수 있게 한다.

• 한국컴퓨터정보학회논문지
• /
• 제23권10호
• /
• pp.23-31
• /
• 2018

Recently, there have been many studies on machine learning. Among them, studies on reinforcement learning are actively worked. In this study, we propose a controller to control bicycle using DDPG (Deep Deterministic Policy Gradient) algorithm which is the latest deep reinforcement learning method. In this paper, we redefine the compensation function of bicycle dynamics and neural network to learn agents. When using the proposed method for data learning and control, it is possible to perform the function of not allowing the bicycle to fall over and reach the further given destination unlike the existing method. For the performance evaluation, we have experimented that the proposed algorithm works in various environments such as fixed speed, random, target point, and not determined. Finally, as a result, it is confirmed that the proposed algorithm shows better performance than the conventional neural network algorithms NAF and PPO.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2022년도 춘계학술발표대회
• /
• pp.161-164
• /
• 2022

Homomorphic Encryption (HE) schemes have been recently growing as a reliable solution to preserve users' information owe to maintaining and operating the user data in the encrypted state. In addition to that, several Neural Networks models merged with HE schemes have been developed as a prospective tool for privacy-preserving machine learning. Those mentioned works demonstrated that it is possible to match the accuracy of non-encrypted models but there is always a trade-off in the computation time. In this work, we evaluate the implementation of CKKS HE scheme operations over the layers of a LeNet5 convolutional inference model, however, owing to the limitations of the evaluation environment, the scope of this work is not to develop a complete LeNet5 encrypted model. The evaluation was performed using the MNIST dataset with Microsoft SEAL (MSEAL) open-source homomorphic encryption library ported version on Python (PyFhel). The behavior of the encrypted model, the limitations faced and a small description of related and future work is also provided.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.148-148
• /
• 2022

Spatiotemporal precipitation data is one of the primary quantities in hydrological as well as climatological studies. Despite the fact that the estimation of these data has made considerable progress owing to advances in remote sensing, the discrepancy between satellite-derived precipitation product (SPP) data and observed data is still remarkable. This study aims to propose an effective deep learning model (DLM) for bias correction of SPPs. In which TRMM (The Tropical Rainfall Measuring Mission), CMORPH (CPC Morphing technique), and PERSIANN-CDR (Precipitation Estimation from Remotely Sensed Information using Artificial Neural Networks) are three SPPs with a spatial resolution of 0.25o exploited for bias correction, and APHRODITE (Asian Precipitation - Highly-Resolved Observational Data Integration Towards Evaluation) data is used as a benchmark to evaluate the effectiveness of DLM. We selected the Mekong River Basin as a case study area because it is one of the largest watersheds in the world and spans many countries. The adjusted dataset has demonstrated an impressive performance of DLM in bias correction of SPPs in terms of both spatial and temporal evaluation. The findings of this study indicate that DLM can generate reliable estimates for the gridded satellite-based precipitation bias correction.

• 한국수자원학회:학술대회논문집
• /
• 한국수자원학회 2022년도 학술발표회
• /
• pp.143-143
• /
• 2022

Precipitation plays an essential role in water resources management and disaster prevention. Therefore, the understanding related to spatiotemporal characteristics of rainfall is necessary. Nowadays, highly accurate precipitation is mainly obtained from gauge observation systems. However, the density of gauge stations is a sparse and uneven distribution in mountainous areas. With the proliferation of technology, satellite-based precipitation sources are becoming increasingly common and can provide rainfall information in regions with complex topography. Nevertheless, satellite-based data is that it still remains uncertain. To overcome the above limitation, this study aims to take the strengthens of machine learning to generate a new reanalysis of precipitation data by fusion of multiple satellite precipitation products (SPPs) with gauge observation data. Several machine learning algorithms (i.e., Random Forest, Support Vector Regression, and Artificial Neural Network) have been adopted. To investigate the robustness of the new reanalysis product, observed data were collected to evaluate the accuracy of the products through Kling-Gupta efficiency (KGE), probability of detection (POD), false alarm rate (FAR), and critical success index (CSI). As a result, the new precipitation generated through the machine learning model showed higher accuracy than original satellite rainfall products, and its spatiotemporal variability was better reflected than others. Thus, reanalysis of satellite precipitation product based on machine learning can be useful source input data for hydrological simulations in ungauged river basins.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2012-2030
• /
• 2021

Named Entity Recognition (NER) is used to identify entity nouns in the corpus such as Location, Person and Organization, etc. NER is also an important basic of research in various natural language fields. The processing of Chinese NER has some unique difficulties, for example, there is no obvious segmentation boundary between each Chinese character in a Chinese sentence. The Chinese NER task is often combined with Chinese word segmentation, and so on. In response to these problems, we summarize the recognition methods of Chinese NER. In this review, we first introduce the sequence labeling system and evaluation metrics of NER. Then, we divide Chinese NER methods into rule-based methods, statistics-based machine learning methods and deep learning-based methods. Subsequently, we analyze in detail the model framework based on deep learning and the typical Chinese NER methods. Finally, we put forward the current challenges and future research directions of Chinese NER technology.

• 한국정보처리학회:학술대회논문집
• /
• 한국정보처리학회 2023년도 추계학술발표대회
• /
• pp.605-607
• /
• 2023

Recent research has highlighted the effectiveness of Deep Learning (DL) techniques in automating the detection of lung sound anomalies. However, the available lung sound datasets often suffer from limitations in both size and balance, prompting DL methods to employ data preprocessing such as augmentation and transfer learning techniques. These strategies, while valuable, contribute to the increased complexity of DL models and necessitate substantial training memory. In this study, we proposed a streamlined and lightweight DL method but effectively detects lung sound anomalies from small and imbalanced dataset. The utilization of 1D dilated convolutional neural networks enhances sensitivity to lung sound anomalies by efficiently capturing deep temporal features and small variations. We conducted a comprehensive evaluation of the ICBHI dataset and achieved a notable improvement over state-of-the-art results, increasing the average score of sensitivity and specificity metrics by 2.7%.

• 대한임베디드공학회논문지
• /
• 제18권3호
• /
• pp.89-100
• /
• 2023

Recently, on-device artificial intelligence (AI) solutions using mobile devices and embedded edge devices have emerged in various fields, such as computer vision, to address network traffic burdens, low-energy operations, and security problems. Although vision transformer deep learning models have outperformed conventional convolutional neural network (CNN) models in computer vision, they require more computations and parameters than CNN models. Thus, they are not directly applicable to embedded edge devices with limited hardware resources. Many researchers have proposed various model compression methods or lightweight architectures for vision transformers; however, there are only a few studies evaluating the effects of model compression techniques of vision transformers on performance. Regarding this problem, this paper presents a performance evaluation of vision transformers on embedded platforms. We investigated the behaviors of three vision transformers: DeiT, LeViT, and MobileViT. Each model performance was evaluated by accuracy and inference time on edge devices using the ImageNet dataset. We assessed the effects of the quantization method applied to the models on latency enhancement and accuracy degradation by profiling the proportion of response time occupied by major operations. In addition, we evaluated the performance of each model on GPU and EdgeTPU-based edge devices. In our experimental results, LeViT showed the best performance in CPU-based edge devices, and DeiT-small showed the highest performance improvement in GPU-based edge devices. In addition, only MobileViT models showed performance improvement on EdgeTPU. Summarizing the analysis results through profiling, the degree of performance improvement of each vision transformer model was highly dependent on the proportion of parts that could be optimized in the target edge device. In summary, to apply vision transformers to on-device AI solutions, either proper operation composition and optimizations specific to target edge devices must be considered.

• 정보과학회 컴퓨팅의 실제 논문지
• /
• 제23권7호
• /
• pp.417-423
• /
• 2017

최근, 딥러닝을 사용 가능한 임베디드 디바이스가 상용화됨에 따라 임베디드 시스템 영역에서도 딥러닝 활용에 대한 다양한 연구가 진행되고 있다. 그러나 임베디드 시스템을 고성능 PC 환경과 비교하면 상대적으로 저사양의 CPU/GPU 프로세서와 메모리를 탑재하고 있으므로 딥러닝 기술의 적용에 있어서 많은 제약이 있다. 본 논문에서는 다양한 최신 딥러닝 네트워크들을 임베디드 디바이스에 적용했을때의 성능을 시간과 전력이라는 관점에서 실험적으로 평가한다. 또한, 호스트 CPU와 GPU 디바이스간의 메모리를 공유하는 임베디드 시스템들의 아키텍처적인 특성을 이용하여 메모리 복사를 줄임으로써 실시간 성능과 저전력성을 높이는 방법을 제시한다. 제안된 방법은 대표적인 공개 딥러닝 프레임워크인 Caffe를 수정하여 구현되었으며, 임베디드 GPU를 탑재한 NVIDIA Jetson TK1에서 성능평가 되었다. 실험결과, 대부분의 딥러닝 네트워크에서 뚜렷한 성능향상을 관찰할 수 있었다. 특히, 메모리 사용량이 높은 AlexNet에서 약 33%의 이미지 인식 속도 단축과 50%의 소비 전력량 감소를 관찰할 수 있었다.

• IEIE Transactions on Smart Processing and Computing
• /
• 제6권3호
• /
• pp.193-199
• /
• 2017

Instead of using only spatial features from a single frame for person re-identification, a combination of spatial and temporal factors boosts the performance of the system. A recurrent neural network (RNN) shows its effectiveness in generating highly discriminative sequence-level human representations. In this work, we implement RNN, three Long Short Term Memory (LSTM) network variants, and Gated Recurrent Unit (GRU) on Caffe deep learning framework, and we then conduct experiments to compare performance in terms of size and accuracy for person re-identification. We propose using GRU for the optimized choice as the experimental results show that the GRU achieves the highest accuracy despite having fewer parameters than the others.