Two papers accepted to two ICML workshops, one as **oral**!
ARLET Workshop (oral), HiLD Workshop
One paper (Gradient Descent with Polyak’s Momentum Finds Flatter Minima via Large Catapults) is accepted to ICML 2024 Workshop: 2nd Workshop on High-dimensional Learning Dynamics (HiLD): The Emergence of Structure and Reasoning. This is joint work with the wonderful undergradudate intern Prin Phunyaphibarn (KAIST Math, equal contributions), my advisor Chulhee Yun (KAIST AI), and collaborators Bohan Wang (USTC) and Huishuai Zhang (Peking University, previously at Microsoft Research Asia - Theory Centre).
Also, another paper (A Unified Confidence Sequence for Generalized Linear Models, with Applications to Bandits) is accepted to ICML 2024 Workshop: Aligning Reinforcement Learning Experimentalists and Theorists (ARLET) as an oral presentation! This is joint work with my advisor Se-Young Yun (KAIST AI) and Kwang-Sung Jun (Univ. of Arizona CS).
My coauthors (Chulhee, Kwang-Sung) will be attending the conference in-person. Catch them if you can in Vienna!
Junghyun Lee
PhD Student
PhD student at GSAI, KAIST, jointly advised by Prof. Se-Young Yun and Prof. Chulhee Yun. Interested in mathematical and theoretical AI, i.e., essentially any machine learning challenges necessitating mathematical analysis. Recently focused on statistical problems arising from RLHF, including interactive machine learning and low-dimensional structure recovery.