Youngjoon Lee

Hi, I’m Youngjoon Lee, a passionate self-taught AI research scientist from Republic of Korea. I’m currently a technical research personnel for alternative military service at Korea Institute for Defense Analyses (KIDA), Republic of Korea. The ongoing research goal is to develop future-oriented core technology and processing technologies for defense data. My recent research interests lie in the field of distributed machine learning and learning theory, with a specific focus on federated learning.

Experience

Education

Skills

• Native in Korean
• Fluent in English

Publication

Journal

· Youngjoon Lee, Jinu Gong, and Joonhyuk Kang, “Federated Transfer Learning via Over-the-Air Computation”, IEEE Commun. Lett., preparing.

· Youngjoon Lee, Sangwoo Park, Jin-Hyun Ahn and Joonhyuk Kang, “Accelerated Federated Learning via Greedy Aggregation”, IEEE Commun. Lett., vol. 26, no. 12, pp. 2919-2923, Dec. 2022.

Conference

· Youngjoon Lee, Sangwoo Park, and Joonhyuk Kang, “Byzantine-Resilient Federated Learning via Reverse Aggregation”, IEEE V. Conf. Commun. (VCC), Virtual, Nov. 2023.

· Youngjoon Lee, Sangwoo Park, and Joonhyuk Kang, “Fast-Convergent Federated Learning via Cyclic Aggregation”, IEEE Int. Conf. Image Process. (ICIP), Kuala Lumpur, Malaysia, Oct. 2023. (Oral)

· Youngjoon Lee, Sangwoo Park, and Joonhyuk Kang, “Security-Preserving Federated Learning via Byzantine-Sensitive Triplet Distance”, arXiv preprint arXiv:2210.16519, Oct. 2022.

· Youngjoon Lee and Joonhyuk Kang, “FedLN: Federated Learning with Local Normalization”, Joint Conf. Commun. Inf. (JCCI), April 2021.

Patent

· “Method and Device for Federated Learning Using Variable Learning Rate”, Youngjoon Lee, Sangwoo Park, Jinu Gong, and Joonhyuk Kang, Patent No. 10-2023-0086433, July 2023.

· “Method and System for Byzantine-Resilient and Personalized Distributed Learning”, Youngjoon Lee, Sangwoo Park, and Joonhyuk Kang, Registration No. 10-2522053, April 2023.

· “Nerual Network Training Method and Appratus Using Federated Learning”, Youngjoon Lee, Sangwoo Park, and Joonhyuk Kang, Registration No. 10-2479793-0000, Dec. 2022.

· “Federated Learning Method and System”, Youngjoon Lee and Joonhyuk Kang, Registration No. 10-2390553-0000, April 2022.

Reference

Award

Scholarship