• Proceedings of the Korean Society of Computer Information Conference
• /
• 2023.07a
• /
• pp.67-68
• /
• 2023

본 논문에서 제안한 Multi Agent Multi Action은 기존의 대화형 시스템 방식인 Single Agent Single Action 구조에 비해 확장성을 갖춘 대화 시스템을 구현하는 방식이다. 시스템을 여러 에이전트로 분할하고, 각 에이전트가 특정 액션에 대한 처리를 담당함으로써 보다 유연하고 효율적인 대화형 시스템을 구현할 수 있으며, 다양한 작업에 특화된 에이전트를 그룹화함으로써 작업의 효율성을 극대화하고, 사용자 경험을 향상 시킬 수 있다.

• Journal of Information Processing Systems
• /
• v.17 no.4
• /
• pp.737-753
• /
• 2021

As the demand for high-quality rendering for mixed reality, videogame, and simulation has increased, global illumination has been actively researched. Monte Carlo path tracing can realize global illumination and produce photorealistic scenes that include critical effects such as color bleeding, caustics, multiple light, and shadows. If the sampling rate is insufficient, however, the rendered results have a large amount of noise. The most successful approach to eliminating or reducing Monte Carlo noise uses a feature-based filter. It exploits the scene characteristics such as a position within a world coordinate and a shading normal. In general, the techniques are based on the denoised pixel or sample and are computationally expensive. However, the main challenge for all of them is to find the appropriate weights for every feature while preserving the details of the scene. In this paper, we compare the recent algorithms for removing Monte Carlo noise in terms of their performance and quality. We also describe their advantages and disadvantages. As far as we know, this study is the first in the world to compare the artificial intelligence-based denoising methods for Monte Carlo rendering.

• Journal of the Korean Society for information Management
• /
• v.41 no.1
• /
• pp.261-282
• /
• 2024

This study proposed and evaluated electroencephalography (EEG)-based and eye-tracking-based methods to determine relevance by utilizing users' implicit relevance feedback while navigating content in a digital library. For this, EEG/eye-tracking experiments were conducted on 32 participants using video, image, and text data. To assess the usefulness of the proposed methods, deep learning-based artificial intelligence (AI) techniques were used as a competitive benchmark. The evaluation results showed that EEG component-based methods (av_P600 and f_P3b components) demonstrated high classification accuracy in selecting relevant videos and images (faces/emotions). In contrast, AI-based methods, specifically object recognition and natural language processing, showed high classification accuracy for selecting images (objects) and texts (newspaper articles). Finally, guidelines for implementing a digital library interface based on EEG, eye-tracking, and artificial intelligence technologies have been proposed. Specifically, a system model based on implicit relevance feedback has been presented. Moreover, to enhance classification accuracy, methods suitable for each media type have been suggested, including EEG-based, eye-tracking-based, and AI-based approaches.

• Journal of Korean Philosophical Society
• /
• v.147
• /
• pp.261-281
• /
• 2018

This paper will cover recent discussions on the risks of human being due to the development of artificial intelligence(AI). We will consider AI research as artificial narrow intelligence(ANI), artificial general intelligence(AGI), and artificial super intelligence(ASI). First, we examine the risks of ANI, or weak AI systems. To maximize efficiency, humans will use autonomous AI extensively. At this time, we can predict the risks that can arise by transferring a great deal of authority to autonomous AI and AI's judging and acting without human intervention. Even a sophisticated system, human-made artificial intelligence systems are incomplete, and virus infections or bugs can cause errors. So I think there should be a limit to what I entrust to artificial intelligence. Typically, we do not believe that lethal autonomous weapons systems should be allowed. Strong AI researchers are optimistic about the emergence of artificial general intelligence(AGI) and artificial superintelligence(ASI). Superintelligence is an AI system that surpasses human ability in all respects, so it may act against human interests or harm human beings. So the problem of controlling superintelligence, i.e. control problem is being seriously considered. In this paper, we have outlined how to control superintelligence based on the proposed control schemes. If superintelligence emerges, it is judged that there is no way for humans to completely control superintelligence at this time. But the emergence of superintelligence may be a fictitious assumption. Even in this case, research on control problems is of practical value in setting the direction of future AI research.

• Journal of the Korea Convergence Society
• /
• v.9 no.8
• /
• pp.185-198
• /
• 2018

This paper was written for the purpose of using as the theoretical basic data of judgment in the situation that there is a growing demand for intelligent design theory to be taught in public schools along with evolution theory. In order to verify the possibility that intelligent design theory, which has little empirical evidence unlike evolutionary theory, could be a scientific theory, what intelligence is and whether the trace of intelligence is actually found in nature was confirmed through literature. If intelligent elements, which are traces of intelligence in nature, are discovered empirically in nature, then intelligent design theory can be recognized as a scientific theory and can also be taught in public schools. The identity and traces of intelligence were found in relevant literature and the traces are found not only in various artificial products derived from human beings such as thinking, knowledge, and civilization, but also in all phenomena of nature. Based on these results, this paper provides a discussion on how the evolutionary theory and intelligent design theory should be handled in the field of school education, as well as how to resolve the conflicts between evolutionary theory and intelligent design theory.

• Journal of the Korea Convergence Society
• /
• v.12 no.10
• /
• pp.203-210
• /
• 2021

This study was conducted to develop a convergence education program using smart farms with both input data(temperature, humidity, etc.) and output data(vegetables, fruits, etc.) that are easily accessible in everyday life so that elementary school students can intuitively and easily understand the principles of artificial intelligence(AI) learning. In order to develop this program, we conducted a prior study analysis of a horticulture, software, robot units in the 2015 Practical Arts curriculum and artificial intelligence education. Based on this, 13 components and 16 achievement criteria were selected, and AI programs of 4 sessions(a total of 8 hours). This program can be used as a reference when developing various teaching materials for artificial intelligence education in the future.

• Journal of the Korean School Mathematics Society
• /
• v.25 no.2
• /
• pp.149-174
• /
• 2022

The purpose of this study is to explore the teaching and learning method using artificial intelligence (AI) in the mathematics classroom. To this end, to predict the direction of mathematics education using AI in the mathematics classroom, this study investigates the fields where AI is applied to education, and discuss issues to consider when introducing AI through scenario development using AI in middle school statistics. This study is meaningful in that it specifically considered how artificial intelligence can be grafted into the mathematics classroom through the development of scenarios that integrate and apply artificial intelligence that has been developed and used segmentally in the current middle school statistics. Afterwards, based on the contents of this study, implications for using AI in the math classroom were derived.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.26 no.7
• /
• pp.892-903
• /
• 2020

The rapid and accurate prediction of storm-surge height during typhoon attacks is essential in responding to coastal disasters. Most methods used for predicting typhoon data are based on numerical modeling, but numerical modeling takes significant computing resources and time. Recently, various studies on the expeditious production of predictive data based on artificial intelligence have been conducted, and in this study, artificial intelligence-based storm-surge height prediction was performed. Several learning data were needed for artificial intelligence training. Because the number of previous typhoons was limited, many synthesized typhoons were created using the tropical cyclone risk model, and the storm-surge height was also generated using the storm surge model. The comparison of the storm-surge height predicted using artificial intelligence with the actual typhoon, showed that the root-mean-square error was 0.09 ~ 0.30 m, the correlation coefficient was 0.65 ~ 0.94, and the absolute relative error of the maximum height was 1.0 ~ 52.5%. Although errors appeared to be somewhat large at certain typhoons and points, future studies are expected to improve accuracy through learning-data optimization.

• Journal of the Korean Society of Systems Engineering
• /
• v.15 no.2
• /
• pp.72-78
• /
• 2019

Pathology is the motor that drives healthcare to understand diseases. The way pathologists diagnose diseases, which involves manual observation of images under a microscope has been used for the last 150 years, it's time to change. This paper is specifically based on tumor detection using deep learning techniques. Pathologist examine the specimen slides from the specific portion of body (e-g liver, breast, prostate region) and then examine it under the microscope to identify the effected cells among all the normal cells. This process is time consuming and not sufficiently accurate. So, there is a need of a system that can detect tumor automatically in less time. Solution to this problem is computational pathology: an approach to examine tissue data obtained through whole slide imaging using modern image analysis algorithms and to analyze clinically relevant information from these data. Artificial Intelligence models like machine learning and deep learning are used at the molecular levels to generate diagnostic inferences and predictions; and presents this clinically actionable knowledge to pathologist through dynamic and integrated reports. Which enables physicians, laboratory personnel, and other health care system to make the best possible medical decisions. I will discuss the techniques for the automated tumor detection system within the new discipline of computational pathology, which will be useful for the future practice of pathology and, more broadly, medical practice in general.

• Korean Journal of Artificial Intelligence
• /
• v.6 no.2
• /
• pp.11-15
• /
• 2018

Breast cancer is one of the leading causes of cancer related death among women. So prediction of overall survival status is important into decided in adjuvant treatment. Deep belief network is a kind of artificial intelligence (AI). We intended to construct prediction model by deep belief network using associated clinicopathologic factors. 103881 cases were found in the Korean Breast Cancer Registry. After preprocessing of data, a total of 15733 cases were enrolled in this study. The median follow-up period was 82.4 months. In univariate analysis for overall survival (OS), the patients with advanced AJCC stage showed relatively high HR (HR=1.216 95% CI: 0.011-289.331, p=0.001). Based on results of univariate and multivariate analysis, input variables for learning model included 17 variables associated with overall survival rate. output was presented in one of two states: event or cencored. Individual sensitivity of training set and test set for predicting overall survival status were 89.6% and 91.2% respectively. And specificity of that were 49.4% and 48.9% respectively. So the accuracy of our study for predicting overall survival status was 82.78%. Prediction model based on Deep belief network appears to be effective in predicting overall survival status and, in particular, is expected to be applicable to decide on adjuvant treatment after surgical treatment.