Sehoon Kim

Ph.D. Candidate in BAIR at UC Berkeley
Contact: sehoonkim at berkeley dot edu
CV: link (latest update: Apr. 2024)
Advisor: Prof. Kurt Keutzer
Research Interests: Efficient AI, HW-SW Codesign, AI Systems

Education

Ph.D. in Computer Science

UC Berkeley, 2020 - Present

B.S. in Electrical and Computer Engineering

Seoul National University, 2015 - 2020

About


I am a 4th year Ph.D. student in Berkeley AI Research (BAIR) at EECS, UC Berkeley. My research spans a full-stack approach to designing efficient and privacy-preserving AI solutions under the supervision of Prof. Kurt Keutzer. More specifically, my research interests lie in:

  • Efficient and hardware-aware model architectures
  • Model compression and optimization techniques for low-cost inference (e.g. quantization and pruning)
  • Algorithms and systems for efficient model training and inference
  • Hardware-software co-design and co-optimization

Before joining UC Berkeley, I was an undergrad majoring in Electrical and Computer Engineering (ECE) at Seoul National University, where I was ranked 1st in the entire class of 2020 (overall GPA: 4.29/4.30, major GPA: 4.30/4.30). In my undergrad years, I was honored to work with Prof. Jangwoo Kim on computer architectures, and with Prof. Byung-Gon Chun on deep learning software systems.

Publications