Hyung-Jin Ko
Sungkyunkwan University · 経済学
研究室紹介
Professor Hyung-Jin Ko's research lab specializes in interdisciplinary research at the intersection of artificial intelligence, finance, and time series analysis. The lab focuses on developing advanced deep learning models for video-to-language understanding, with an emphasis on end-to-end trainable concept detection and semantic attention mechanisms to enhance language generation. In parallel, the lab investigates financial applications of emerging digital assets such as NFTs and DeFi tokens, analyzing their role as hedges and safe havens in global markets. The lab also develops adaptive ensemble learning methods for time-series data, particularly in dynamic and non-stationary environments.
Research Overview
Research Output Trend
Figures are computed from collected data and may differ slightly.
Selected Papers
15We propose a high-level concept word detector that can be integrated with any video-to-language models. It takes a video as input and generates a list of concept words as useful semantic priors for language generation models. The proposed word detector has two important properties. First, it does not require any external knowledge sources for training. Second, the proposed word detector is trainable in an end-to-end manner jointly with any video-to-language models. To effectively exploit the det
We propose a high-level concept word detector that can be integrated with any video-to-language models. It takes a video as input and generates a list of concept words as useful semantic priors for language generation models. The proposed word detector has two important properties. First, it does not require any external knowledge sources for training. Second, the proposed word detector is trainable in an end-to-end manner jointly with any video-to-language models. To maximize the values of dete
This study conducted the econometric analysis to test the hedge and safe haven effects of Non-fungible Tokens (NFTs) on major traditional asset markets in the global financial system. We investigate the estimates of these effects in times of extreme market conditions and the COVID-19 crisis. Our empirical results show evidence of the hedge and safe haven properties of NFTs, confirming two main findings: (i) NFTs act as a hedge and safe haven for particular stock markets and oil, bond, and USD in
We propose a high-level concept word detector that can be integrated with any video-to-language models. It takes a video as input and generates a list of concept words as useful semantic priors for language generation models. The proposed word detector has two important properties. First, it does not require any external knowledge sources for training. Second, the proposed word detector is trainable in an end-to-end manner jointly with any video-to-language models. To maximize the values of dete
In decentralized finance (Defi), market participants are allowed to have the right to manage their own funds as opposed to centralized finance (Cefi) with a central custodian, centralized exchanges (CEX). Most Defi projects provide their own service and simultaneously issue a unique token that can be traded in decentralized exchanges (DEX). However, the values of these tokens have rarely been studied. We confirm that the prices of tokens in the Defi market have a persistent tendency to move toge
Developing a robust and sustainable system is an important problem in which deep learning models are used in real-world applications. Ensemble methods combine diverse models to improve performance and achieve robustness. The analysis of time series data requires dealing with continuously incoming instances; however, most ensemble models suffer when adapting to a change in data distribution. Therefore, we propose an on-line ensemble deep learning algorithm that aggregates deep learning models and