Research Projects

Current Projects (* indicates PI) [ML] Machine Learning, [FL] Federated Learning, [MM] Multimodal, [CV] Computer Vision, [AIX] AI+Application

[ML] [MM] Development of a Virtual Tactile Signal Generation Platform Technology Based on Multimodal Vision-Tactile Integrated AGI (Jul. 2025 - Dec. 2028)

• Funded by IITP

• This research project aims to develop vision-tactile integrated AGI. We focus on developing 3D visual reconstruction methods integrated with tactile signals.

[FL] [AIX] FLAME-ARK: Federated Learning Multimodal Foundation Models for Medical Institutes (Apr. 2025 - Dec. 2029)

• Funded by the Korean Ministry of Health and Welfare

• This research project aims to verify federated learning solutions for establishing multimodal foundational models across multiple medical institutes.

[ML] *NRF 개인기초연구 중견연구 과제(Sep. 2024 - Aug. 2027)

• Funded by NRF

• This research project is for developing continual learning of multimodal generative models, including the following issues: i) Unravelling how catastrophic forgetting occurs in generative models, ii) Explaining the synergetic effects between different modalities, and iii) Developing effective continual learning algorithms for multimodal deep generative models.

[ML] AI Star Fellowship Program (Jul. 2025 - Dec. 2030)

• Funded by IITP

• Participated in the program as the Project Leader (PL) of PROJECT 3 (among three core projects in the program)

• The overarching objective is to develop a robust VLA integrated on-device AI for manufacturing industries. The theme of PROJECT 3 to be led by Prof. Yoon is to develop a robust, reliable, and explainable reinforcement learning for optimization.

[ML] Graduate School of Artificial Intelligence (인공지능대학원) in UNIST (Apr. 2020 - present) - [link]

• • • Funded by IITP

    • Participated as a core professor

[AIX] 5T Space UNIST(지역지능화 혁신인재 양성과제, Jan. 2023 - Dec. 2029)

• Funded by IITP

Past Projects(* indicates PI)

*ETRI 위탁과제(다계층 연합학습 시뮬레이션 및 기여도 측정 도구 제작, Jun. 2024 - Nov. 2023)

• Funded by ETRI

• This research project is to develop a robust and flexible simulation environment for hierarchical and asynchronous federated learning algorithms.

한국-미국(NSF) 국립과학재단 국제 공동 연구(Dec. 2021 - Nov. 2024)

• Funded by IITP

• This research project is for developing privacy-preserving video caching algorithms based on federated learning.

• Collaboration with USC (Prof. A. Molisch), Kyunghee Univ. (Prof. Minseok Choi), and Inha Univ. (Prof. Yeongjin Kim)

• Academic results: 1 paper for domain generalization in AAAI 2023, 1 paper for multi-domain meta-learning in AISTATS 2024, 2 paper for federated learning in WACV 2024, ICML 2024 1 paper for intelligent communication in WoWMoM 2024, and 1 workshop paper in AAAI'23 workshop.

*NRF 개인기초연구 우수신진연구 과제(Mar. 2021 - Feb. 2024)

• Funded by NRF (National Research Foundation)

• This research project is for developing context-aware self-growing continual meta-learning with a few training data samples.

• Academic results: 1 paper for generalization in AAAI 2023, 1 paper for few-shot class-incremental learning in WACV 2024, 1 paper for multi-domain meta-learning in AISTATS 2024.

*ETRI 위탁과제(다중 추론 성능 기반 연합학습 비동기 합의 실험 환경 구축, Jun. 2023 - Nov. 2023)

• Funded by ETRI

• This research project is for developing a hierarchical and asynchronous federated learning algorithm.

*ETRI 위탁과제(자원 은닉형 딥러닝 모델 실현 가능성 검증 기술 개발, Apr. 2023 - Nov. 2023)

• Funded by ETRI

• This research project is for developing double-blind federated learning algorithms that secure both models and data.

ITRC 대학 정보통신기술 연구센터(Jan. 2021 - Dec. 2022)

• Funded by IITP

• This research project is for developing machine learning algorithms for energy-efficient wireless sensor management.

• PI: Prof. Franklin Bien (UNIST EE)

Mobility-On-Demand AI Technologies for Autonomous Vehicles (Mar. 2021 - Feb. 2023)

• Funded by UNIST and (주) 씨엘

Multi-objective Neural Combinatorial Optimization Algorithm (Mar. 2021 - Feb 2023)

• Funded by UNIST and EpiSci

*LG H&A 산학 연구 과제(Jul. 2022 - Jan. 2023)

• Funded by LG 전자 H&A

*고효율/고신뢰 연소 시스템을 위한 실 데이터 기반 강화 학습 솔루션 개발 (Jan. 2022 - Dec. 2022)

• Funded by 삼양사(주) & UNIST

• This research project is for developing high-efficient and reliable combustion control systems based on deep reinforcement learning algorithms.

*Machine Learning for Error-Correcting Codes (May 2020 - Nov. 2020)*

• • • Funded by, ETRI

    • Our research part is focused on assessing the possibility to merge machine learning algorithms and error-correcting codes together.

KAIST-SK AI Research Center (2019 - 2020)

• • • Funded by, SK hynix Inc.

    • My research part is focused on developing key machine learning algorithms for automated vehicle system including semantic segmentation and multi-modal learning.

Research on Adaptive Machine Learning Technology Development for Intelligent Autonomous Digital Companion (2016 - 2020) [Link]

• • • Funded by, ICT R&D program of Institute for Information & Communications Technology Promotion (IITP) grant funded by the Korea government (MSIP)

    • My research part is focused on developing meta-learning algorithms for an intelligent autonomous digital companion. As results, the meta-learner with linear nulling is proposed.

    • Academic results: 1 paper for few-shot learning in ICML 2019, 1 workshop paper in meta-learning workshop at NIPS 2018 (NeurIPS)

Cloud Storage Coding (2016 - 2019)

• • • Funded by,  National Research Foundation of Korea government

    • My research part is focused on mathematical analysis of storage capacity & security issue in clustered distributed storage system (partial contribution as a co-author).

    • Academic results: 1 conf. paper for the security issue / 1 journal & 1 conf. paper for the analysis on the capacity of clustered distributed storage

KAIST-SK hynix Storage Media Solutions Center (2013 - 2020)

• • • Funded by, SK hynix Inc.

    •  My research part is focused on the high-performance and low-complexity algorithm/hardware of error correcting codes for NAND ash memory.

    • Academic results: 1 conf. paper for low-complexity concatenated polar coding algorithm for storage