• The Journal of the Convergence on Culture Technology
• /
• v.10 no.5
• /
• pp.645-650
• /
• 2024

This paper jointly designed and developed a liberal arts curriculum at two local universities for college liberal arts education using generative AI centered on ChatGPT. The developed curriculum takes into account the conceptual components for designing classes for integrated use of university ChatGPT presented in existing research, understands the language model and artificial intelligence that form the basis of ChatGPT, and applies generative AI including ChatGPT to various domains. It was developed with useful content. The developed curriculum introduces the concept and changing aspects of artificial intelligence and the natural language processing language model that is the basis of ChatGPT for students in various majors, and generates ChatGPT, a generative AI and large language model (LLM), and various open sources. The purpose was to implement my own AI service using the model and present an example of mutual collaboration between universities in Joint Education Curriculum Operation.

• Annual Conference on Human and Language Technology
• /
• 2023.10a
• /
• pp.409-414
• /
• 2023

최근 지식 기반 대화 생성에 많은 연구자가 초점을 맞추고 있다. 특히, 특정 도메인에서의 작업 지향형 대화 시스템을 구축하는 것은 다양한 도전 과제가 있으며, 이 중 하나는 거대 언어 모델이 입력과 관련된 지식을 활용하여 응답을 생성하는 데 있다. 하지만 현재 거대 언어 모델은 작업 지향형 대화에서 단순히 정보를 열거하는 방식으로 응답을 생성하는 경향이 있다. 이 논문에서는 전문 지식과 대화 정책 예측 모델을 결합한 프롬프트를 제시하고 작업 지향형 대화에서 사용자의 최근 입력에 대한 정보 제공 및 일상 대화를 지원하는 가능성을 탐구한다. 이러한 새로운 접근법은 모델 파인튜닝에 비해 비용 측면에서 효율적이며, 향후 대화 생성 분야에서 발전 가능성을 제시한다.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.84-86
• /
• 2022

최근 미디어, 금융 등 다양한 분야의 기업들이 AI를 활용해 제공하는 서비스가 늘어남에 따라 학습된 모델을 엣지 자원에 배포하여 기능을 제공하는 서비스형태 또한 늘어나고 있다. AI-Application이 동작하기 위해서는 AI-Model 파일뿐 아니라 동작을 위한 설정 파일들이 필요하여 AI-Application이 사용 중인 AI-Model의 정보를 수집, 관리하는 것은 중요한 이슈라고 할 수 있다. 하지만 단일 서비스서버에서 동작하는 형태가 아닌 각 자원이 산재되어 다양한 형태로 서비스를 제공하는 엣지컴퓨팅의 구조적인 특성상 AI-Application의 기존 서비스구조, 기능을 수정하지 않고 정보를 수집하는 과정은 다양한 문제에 부딪치게 된다. 이에 따라 본 논문에서는 기존 서비스구조를 변경하지 않고 독립적으로 AI-Application에서 사용중인 AI-Model의 정보를 파악하고, 사용자 요청에 대응할 수 있는 관리구조를 제안한다.

• Journal of Intelligence and Information Systems
• /
• v.28 no.2
• /
• pp.191-206
• /
• 2022

Recently, it is a de-facto approach to utilize a pre-trained language model(PLM) to achieve the state-of-the-art performance for various natural language tasks(called downstream tasks) such as sentiment analysis and question answering. However, similar to any other machine learning method, PLM tends to depend on the data distribution seen during the training phase and shows worse performance on the unseen (Out-of-Distribution) domain. Due to the aforementioned reason, there have been many efforts to develop domain-specified PLM for various fields such as medical and legal industries. In this paper, we discuss the training of a finance domain-specified PLM for the Korean language and its applications. Our finance domain-specified PLM, KB-BERT, is trained on a carefully curated financial corpus that includes domain-specific documents such as financial reports. We provide extensive performance evaluation results on three natural language tasks, topic classification, sentiment analysis, and question answering. Compared to the state-of-the-art Korean PLM models such as KoELECTRA and KLUE-RoBERTa, KB-BERT shows comparable performance on general datasets based on common corpora like Wikipedia and news articles. Moreover, KB-BERT outperforms compared models on finance domain datasets that require finance-specific knowledge to solve given problems.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_2
• /
• pp.1581-1587
• /
• 2022

Recently, deep learning has become more important in remote sensing data processing. Huge amounts of data for artificial intelligence (AI) has been designed and built to develop new technologies for remote sensing, and AI models have been learned by the AI training dataset. Artificial intelligence models have developed rapidly, and model accuracy is increasing accordingly. However, there are variations in the model accuracy depending on the person who trains the AI model. Eventually, experts who can train AI models well are required more and more. Moreover, the deep learning technique enables us to automate methods for remote sensing applications. Methods having the performance of less than about 60% in the past are now over 90% and entering about 100%. In this special issue, thirteen papers on how deep learning techniques are used for remote sensing applications will be introduced.

• Smart Media Journal
• /
• v.12 no.9
• /
• pp.159-170
• /
• 2023

It takes a lot of time and manpower to search for the missing. As part of the solution, a missing person search AI system was implemented using a YOLO-based model. In order to train object detection models, the model was learned by collecting recognition images (road fixation) of drone mobile objects from AI-Hub. Additional mountainous terrain datasets were also collected to evaluate performance in training datasets and other environments. In order to optimize the missing person search AI system, performance evaluation based on model size and hyperparameters and additional performance evaluation for concerns about overfitting were conducted. As a result of performance evaluation, it was confirmed that the YOLOv5-L model showed excellent performance, and the performance of the model was further improved by applying data augmentation techniques. Since then, the web service has been applied with the YOLOv5-L model that applies data augmentation techniques to increase the efficiency of searching for missing people.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.21 no.5
• /
• pp.71-77
• /
• 2021

Recently, there is a demand for efficient program development of an IoT application support frameworks considering heterogeneous hardware characteristics. In addition, the scope of hardware support is expanding with the development of neuromorphic architecture that mimics the human brain to learn on their own and enables autonomous computing. However, most existing IoT IDE(Integrated Development Environment), it is difficult to support AI(Artificial Intelligence) or to support services combined with various hardware such as neuromorphic architectures. In this paper, we design an AI component abstract model that supports the second-generation ANN(Artificial Neural Network) and the third-generation SNN(Spiking Neural Network), and implemented an autonomous IoT IDE based on the proposed model. IoT developers can automatically create AI components through the proposed technique without knowledge of AI and SNN. The proposed technique is flexible in code conversion according to runtime, so development productivity is high. Through experimentation of the proposed method, it was confirmed that the conversion delay time due to the VCL(Virtual Component Layer) may occur, but the difference is not significant.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.01a
• /
• pp.457-460
• /
• 2023

본 연구는 시설작물의 병충해 진단을 위해 딥러닝 모델을 응용한 인공지능 서비스 앱, Dr. Vegetable을 제안하고자 한다. 농업 현장에서 숙련된 농부는 한눈에 농작물의 병충해를 판단할 수 있지만 미숙련된 농부는 병충해 피해를 발견하더라도 그 종류와 해결 방법을 찾아내기가 매우 어렵다. 또한 아무리 숙련된 농부라고 할지라도 육안검사만으로 병충해를 조기에 발견하는 것은 쉽지 않다. 한편 시설작물의 경우 병충해에 의한 연쇄피해가 발생할 우려가 있으므로 병충해의 조기 발견 및 방제가 매우 중요하다. 즉, 농부의 경험에 따른 농작물 병해충 진단은 정확성을 장담할 수 없으며 비용과 시간적인 측면에서 위험성이 높다고 할 수 있다. 본 논문에서는 YOLOv5를 활용하여 상추, 고추, 토마토 등 농작물의 병충해를 진단하는 인공지능 서비스를 제안한다. 특히 한국지능정보사회진흥원이 운영하고 있는 AI 통합 플랫폼인 AI 허브에서 제공하는 노지 작물 질병 및 해충 진단 이미지를 사용하여 딥러닝 모델을 학습하였다. 본 연구를 통해 개발된 모바일 어플리케이션을 이용하여 실제 시설농장에서 병충해 진단 서비스를 적용한 결과 약 86%의 정확도, F1 Score 0.84, 그리고 0.98의 mAP 값을 얻을 수 있었다. 본 연구에서 개발한 병충해 진단 딥러닝 모델을 다양한 조도에서 강인하게 동작하도록 개선한다면 농업 현장에서 널리 활용될 수 있을 것으로 기대한다.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.32 no.5
• /
• pp.817-826
• /
• 2022

AI technology is being successfully introduced in many fields, and models deployed as a service are deployed with black box environment that does not expose the model's information to protect intellectual property rights and data. In a black box environment, attackers try to steal data or parameters used during training by using model output. This paper proposes a method of inferring the type of model to directly find out the composition of layer of the target model, based on the fact that there is no attack to infer the information about the type of model from the deep learning model. With ResNet, VGGNet, AlexNet, and simple convolutional neural network models trained with MNIST datasets, we show that the types of models can be inferred using the output values in the gray box and black box environments of the each model. In addition, we inferred the type of model with approximately 83% accuracy in the black box environment if we train the big and small relationship feature that proposed in this paper together, the results show that the model type can be infrerred even in situations where only partial information is given to attackers, not raw probability vectors.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.5
• /
• pp.827-832
• /
• 2024

The AI program Sora is a video production model that can be used innovatively and is the starting point of a major paradigm shift in video planning and production in the future. In this paper, through consideration of the characteristics, application, and process of the AI video production program, the characteristics of the AI design video production method were understood, and the production algorithm was considered. The detailed consideration and characteristics of the work creation process for the video graphic AI video generation program that will be intensified every year were examined. Next, the method of generating a customized video with a text prompt and the process of innovative production results different from the previous production method were considered. In addition, the design direction through the generation of AI images was studied through the review of the strengths and weaknesses of the image details of the recently announced AI music video results. By considering the security of the AI generation video Sora and looking at the internal process of the actual AI process, it will be possible to present indicators for the future direction of AI video model production and education along with the direction of the design designer and education system. In the text and conclusion, we analyzed the strengths and weaknesses and future status of OpenAI Sora image, concluded how to apply the Sora model's capabilities, limitations, quality, and human creativity, and presented problems and alternatives through examples of the Sora model's capabilities and limitations to increase human creativity.