Chulhee Yun

Chulhee Yun is an Associate Professor at the Kim Jaechul Graduate School of AI, KAIST, where he also holds a joint affiliation with the KAIST Graduate School of AI for Math and serves as a part-time Visiting Faculty Researcher at Google Research. He directs the Optimization & Machine Learning (OptiML) Laboratory at KAIST AI, focusing on the theoretical principles of optimization and deep learning. His research bridges mathematical theory with practical neural network training. Yun received his PhD in Electrical Engineering and Computer Science from MIT, where he was jointly supervised by Prof. Suvrit Sra and Prof. Ali Jadbabaie, and his academic journey includes a master’s in Electrical Engineering from Stanford University and a bachelor’s in Electrical Engineering from KAIST. His work has been recognized with several awards, including the KT Best Paper Award at the KAIA Summer Conference 2025 and the KAIA Outstanding Paper Award at the KAIA Summer Conference 2025.

Chulhee Yun is an Associate Professor at KAIST Kim Jaechul Graduate School of AI, specializing in Optimization, Deep Learning Theory, and Machine Learning Theory. His research explores the theoretical foundations of deep learning, focusing on approximation capabilities, optimization dynamics, and generalization. Key contributions include analyzing the universality of transformers, understanding the role of nonlinearities in neural networks, and studying the impact of shuffling in stochastic gradient descent. His work also investigates memorization capacity, implicit bias, and the interplay between model structure and performance. Yun's research bridges theoretical insights with practical implications for training and deploying deep learning models.