Chulhee Yun

Chulhee Yun is an Ewon Assistant Professor at KAIST Kim Jaechul Graduate School of AI (KAIST AI), with a joint affiliation starting September 2025 at KAIST Graduate School of AI for Math and a part-time Visiting Faculty Researcher position at Google Research. He directs the Optimization & Machine Learning (OptiML) Laboratory at KAIST AI.

Education:

• PhD in Elec. Eng. & Comp. Sci., 2016–2021, Massachusetts Institute of Technology
• MSc in Electrical Engineering, 2014–2016, Stanford University
• BSc in Electrical Engineering, 2007–2014, KAIST

Chulhee Yun's research focuses on deep learning theory, optimization, and machine learning theory. His work bridges theoretical foundations with practical applications in neural network training, addressing fundamental questions about optimization dynamics, generalization capabilities, and the mathematical properties of deep learning models. His research spans from theoretical analyses of gradient-based optimization methods to practical techniques for improving neural network training and performance.

His recent publications demonstrate a strong focus on understanding the fundamental properties of neural network training dynamics, optimization algorithms, and generalization. The research spans theoretical analyses of stochastic optimization methods like SGD, investigations into the properties of transformers and language models, and practical techniques for improving neural network training efficiency. A notable trend is the combination of rigorous theoretical analysis with practical implications for deep learning systems.

Scientific Awards:

• KAIA Outstanding Paper Award at KAIA Summer Conference 2025 (CKAIA 2025)
• KT Best Paper Award at KAIA Summer Conference 2025 (CKAIA 2025)
• Best Paper Award at KAIA Fall Conference 2024 (JKAIA 2024)
• KT Best Paper Award at KAIA Summer Conference 2024 (CKAIA 2024)
• KAIA Outstanding Paper Award at KAIA Summer Conference 2023 (CKAIA 2023)
• NAVER Outstanding Theory Paper Award at KAIA-NAVER Joint Fall Conference 2022 (JKAIA 2022)
• Honorable Mention at NYAS Machine Learning Symposium 2020 Poster Awards
• Notable AC for NeurIPS 2023
• Top Reviewer at ICML 2025

Professor Yun actively mentors PhD, Master's, and undergraduate students through the OptiML Laboratory. His research group includes several PhD students (Jaeyoung Cha, Hanseul Cho, Yujun Kim, Junghyun Lee, Junsoo Oh, Baekrok Shin), Master's students, and undergraduate interns. He has successfully guided former students such as Geonhui Yoo, Jaewook Lee, and Dongkuk Si to completion. His service to the academic community includes serving as Social Chair for ICML 2026, Conference Area Chair for ICLR 2025-2026 and NeurIPS 2023-2025, and as a reviewer for numerous top conferences and journals.

Chulhee Yun directs the Optimization & Machine Learning (OptiML) Laboratory at KAIST, which focuses on advancing the theoretical foundations of machine learning and optimization. The lab investigates fundamental questions about neural network training dynamics, optimization algorithms, and generalization properties, with the goal of bridging theoretical insights with practical applications in deep learning systems.