Seyeon Kim is currently a Post Doctoral Researcher in the Department of Electrical and Computer Engineering at Seoul National University. He received his B.S., M.S., and Ph.D. degrees in Electrical Engineering from KAIST (Korea Advanced Institute of Science and Technology), Daejeon, Korea, in 2015, 2017, and 2022, respectively. His research interests include thermally reliable and energy-efficient mobile deep learning systems, neural network model compression, vertical/horizontal split computing for neural networks, and offloaded analytics for AR/XR/MR. He is the recipient of the best paper award of ACM MobiSys 2021 for his ground-breaking work, zTT which proposed the first application- and environment-aware DVFS with reinforcement learning for achieving zero thermal throttling in mobile systems. 

Current research interest: Near-memory networking/computing, Thermally-reliable wireless networking, Real-time (volumetric) video streaming, AI-based cloud management

• B.S., EE, KAIST, South Korea, 2015

• M.S., EE, KAIST, South Korea, 2017 (Advisor: Prof. Song Chong (KAIST))

• Ph.D., EE, KAIST, South Korea, 2022 (Advisor: Prof. Song Chong (KAIST), Co-advised by Prof. Kyunghan Lee (SNU))

• Postdoctoral Researcher, Networked Computing and AI Lab (Host: Prof. Kyunghan Lee), ECE, Seoul National University, South Korea, 2022.09 - 2023.12

• Postdoctoral researcher, Internet Systems Lab (Host: Prof. Sangtae Ha), CS, University of Colorado at Boulder, USA, 2024.02 - Present

Academia

• IEEE MASS'24 (TPC Member)

• ACM Mobisys Best Paper Award 2021

Awarded for the paper: "zTT: Learning-based DVFS with Zero Thermal Throttling for Mobile Devices"

• On Designing a Thermally-reliable Mobile System (A3 Foresight Workshop for AI-based Future IoT Technologies and Services, Tokyo, Japan, 2022.12.19)

• On Learning-based Mobile Performance Guarantee under Time-varying Resource Constraints (DIGIST, 2022.08.26)

• 강화 학습의 기초 및 응용 (머신러닝/강화학습의 기초 및 응용 강좌, KICS, Online, 2022.07.04)

• zTT: Learning-based DVFS with Zero Thermal Throttling for Mobile Devices (A3 Foresight Workshop, Online,  2021.08.11)

• 심층 강화 학습과 응용 사례 (제1회 강화학습 기초 및 응용 강좌, KICS, Online, 2020.08.24)

• To be submitted: ACM NSDI'25 (1), ACM/IEEE Micro'24 (1), ACM CoNext'24 (1)

• Under review: ACM Mobicom'24 (1), IEEE TMC (1)

• To be presented: ACM Mobisys'24 (Accepted), IEEE DySPAN'24 (Accepted), IJCAI'24 (Accepted)

• Seyeon Kim, Kyungmin Bin, Donggyu Yang, Sangtae Ha, Kyunghan Lee*, Song Chong. “ENTRO: Tackling the Encoding and Networking Trade-off in Offloaded Video Analytics.” Proceedings of the 31th ACM International Conference on Multimedia. 2023. 

• Insoo Lee, Seyeon Kim, Sandesh Dhawaskar Sathyanarayana, Kyungmin Bin, Song Chong, Kyunghan Lee*, Drik Grunwald, Sangtae Ha*. “RL-based FEC Adjustment for better QoE in WebRTC.” Proceedings of the 30th ACM International Conference on Multimedia. 2022.

• Seyeon Kim, Kyungmin Bin, Sangtae Ha, Kyunghan Lee, Song Chong. “zTT: Learning-based DVFS with zero thermal throttling for mobile devices.” GetMobile: Mobile Computing and Communications 25.4 (2022): 30-34. (Invited paper)

• Seyeon Kim, Kyungmin Bin, Sangtae Ha, Kyunghan Lee*, Song Chong. “zTT: Learning-based DVFS with zero thermal throttling for mobile devices.” Proceedings of the 19th Annual International Conference on Mobile Systems, Applications, and Services. 2021.(ACM Mobisys 2021)