Faculty Research Profile
Jinyoung Choi
학력
· 2025: Ph.D., Electrical and Computer Engineering, Seoul National University
· 2012: M.S., Mathematics, Pohang University of Science and Technology
· 2010: B.S., Industrial and Management Engineering & Mathematics (Double Major), Pohang University of Science and Technology
주요 경력
· 2026-present: Assistant Professor, UNIST
· 2025-2026: PostDoc., AI Institute, Seoul National University
· 2012-2018: Research Engineer~Senior Research Engineer, LG Production Engineering Research Institute, LG Electronics
수상/학회/외부활동
· 2025: Gold Prize, Samsung Humantech Paper Awards Feb 2025
Samsung Electronics
생성 지능 연구실
Generative Intelligence Lab
생성 지능 연구실은 생성 모델이 현대 인공지능의 모든 과업을 구동하는 핵심 기반 기술이라는 관점에서 데이터의 근본적인 생성 원리를 탐구한다. 수학적 이론을 토대로 데이터 내의 복잡한 패턴을 깊이 있게 이해하고 새로운 가치를 창조하는 방법론을 설계한다. 나아가 이러한 고도화된 생성 기술을 산업 및 공학 분야의 실질적인 복잡한 문제 해결에 적용하는 것을 목표로 한다.
The Generative Intelligence Lab explores the fundamental generative principles of data based on the perspective that generative models serve as the core engine powering all tasks in modern artificial intelligence. We design original methodologies rooted in mathematical theory to deeply understand complex patterns in data and create new value. Further, we aim to apply these advanced generative technologies to solve practical challenges in industrial and engineering domains.
The Generative Intelligence Lab explores the fundamental generative principles of data based on the perspective that generative models serve as the core engine powering all tasks in modern artificial intelligence. We design original methodologies rooted in mathematical theory to deeply understand complex patterns in data and create new value. Further, we aim to apply these advanced generative technologies to solve practical challenges in industrial and engineering domains.
연구분야
생성 모델, 딥러닝, 머신러닝, 컴퓨터 비전 / Generative Models, Deep Learning, Machine Learning, Computer Vision
Generative Models, Deep Learning, Machine Learning, Computer Vision
연구주제
디퓨전 확률 모델, 스코어 기반 생성 모델, 확률 미분 방정식, 적대적 학습, 매니폴드 학습, 표현 학습, 추론 가속화
Diffusion Probabilistic Models, Score-based Generative Models, Stochastic Differential Equations (SDEs), Adversarial Learning, Manifold Learning, Representation Learning, Inference Acceleration
Diffusion Probabilistic Models, Score-based Generative Models, Stochastic Differential Equations (SDEs), Adversarial Learning, Manifold Learning, Representation Learning, Inference Acceleration
국가연구개발사업 기술 분류체계
국가과학기술표준분류
EE. 정보/통신 > EE02. 소프트웨어 > EE0299. 달리 분류되지 않는 소프트웨어
논문
· The Thirteenth International Conference on Learning Representations (ICLR), Enhanced Diffusion Sampling via Extrapolation with Multiple ODE Solutions, Jinyoung Choi / Junoh Kang / Bohyung Han. (2025)
· The IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), Observation-Guided Diffusion Probabilistic Models, Junoh Kang* / Jinyoung Choi* / Sungik Choi / Bohyung Han. (2024)
· The Thirty-Sixth Annual Conference on Neural Information Processing Systems (NeurIPS), MCL-GAN: Generative Adversarial Networks with Multiple Specialized Discriminators, Jinyoung Choi / Bohyung Han. (2022)
특허
[국내] 타겟 태스크별 양자화 테이블 생성 방법, 한보형/최진영(2023.2 등록)
[국내] 선입선출 디퓨전을 이용한 텍스트 기반의 무한 비디오 생성 방법 및 시스템, 김지환/강준오/최진영/한보형(2024.12 출원)