• Juno Kim, Taiji Suzuki. Transformers Provably Solve Parity Efficiently with Chain of Thought. NeurIPS 2024 Workshop on Mathematics of Modern Machine Learning. arxiv:2410.08633

• Juno Kim, Dimitri Meunier, Arthur Gretton, Taiji Suzuki, Zhu Li. Optimality and Adaptivity of Deep Neural Features for Instrumental Variable Regression. NeurIPS 2024 Workshop on Mathematics of Modern Machine Learning.

• Juno Kim, Tai Nakamaki, Taiji Suzuki. Transformers are Minimax Optimal Nonparametric In-Context Learners. NeurIPS 2024 & ICML 2024 Workshop on Theoretical Foundations of Foundation Models (Best Paper Award). arXiv:2408.12186 

• Juno Kim, Taiji Suzuki. Transformers Learn Nonlinear Features In Context: Nonconvex Mean-field Dynamics on the Attention Landscape. ICML 2024 (oral presentation). arXiv:2402.01258

• Juno Kim, Kakei Yamamoto, Kazusato Oko, Zhuoran Yang, Taiji Suzuki. Symmetric Mean-field Langevin Dynamics for Distributional Minimax Problems. ICLR 2024 (spotlight paper). arXiv:2312.01127

• Juno Kim, Jaehyuk Kwon, Mincheol Cho, Hyunjong Lee, Joong-Ho Won. t3-Variational Autoencoder: Learning Heavy-tailed Data with Student's t and Power Divergence. ICLR 2024. arXiv:2312.01133

• Juno Kim, Otto van Koert. Hessian Based Smoothing Splines for Manifold Learning. 2023, arXiv preprint arXiv:2302.05025

• Juno Kim, Yonghwan Kim, Otto van Koert. Reeb Flows without Simple Global Surfaces of Section. Involve, 15(5), pp. 813–842, 2022. arXiv:2104.03728

RIKEN Center for Advanced Intelligence Project, Tokyo, Japan Dec 2023 – present
Deep Learning Theory Team, Part-time Researcher

University of Tokyo, Tokyo, Japan Apr 2023 – present
M.Sc. in Mathematical Informatics (expected graduation date: Mar 2025)

Seoul National University, Seoul, South Korea Mar 2018 – Feb 2023
B.Sc. in Mathematical Sciences
B.Sc. in Statistics
Graduated Valedictorian of the College of Natural Sciences (GPA 4.28/4.3)
Undergraduate Research Intern Jun 2019 – Feb 2023
Dept. of Statistics Peer Tutor (Mathematical Statistics I, II) Mar 2022 – Dec 2022
Military Service Sep 2020 – Mar 2022

Seoul Science High School, Seoul, South Korea Mar 2015 – Feb 2018

Scholarships
Doctoral Course (DC1) Research Fellowship, JSPS selected
Japanese Government Scholarship Apr 2023 – present
National Scholarship, Kwanjeong Educational Foundation Mar 2020 – Feb 2023
Eminence Scholarship, Seoul National University Sep 2018 – Feb 2020

Awards
President Award, Highest Honors, Seoul National University Feb 2023
President Award, Korean Statistical Society Feb 2023
4th Place, Simon Marais Mathematics Competition Oct 2022
Gold Prize, College Mathematics Competition Dec 2019 

I love attending conferences, traveling to new places and meeting people who share my interests. My hobbies are weight training, recreational math and playing the guitar. I am fluent in Korean, English and Japanese and learning German and French.

2024
Neural Information Processing Systems, Vancouver,  Canada Dec 9 – 16
Simons Institute, Berkeley (CA), USA (visiting) Nov 11 – Dec 9
UCL Gatsby Computational Neuroscience Unit, London, UK (visiting) Aug 5 – 19
International Conference on Machine Learning, Vienna, Austria Jul 21 – 27
International Conference on Learning Representations, Vienna, Austria May 7 – 11
UCL Gatsby Computational Neuroscience Unit, London, UK (visiting) Mar 28
Workshop on Functional Inference and Machine Intelligence, Bristol, UK Mar 25 – 27
Tokyo Deep Learning Workshop, Tokyo, Japan Mar 18 – 20
Machine Learning Summer School, Okinawa, Japan Mar 4 – 15

2023
NYU Center for Data Science, New York, USA (visiting) Dec 16 – 19
Neural Information Processing Systems, New Orleans, USA Dec 10 – 16
Information-Based Induction Sciences Workshop, Kyushu, Japan Oct 29 – Nov 1
Japanese Joint Statistical Meeting, Kyoto, Japan Sep 3 – 7