I’m a Ph.D. student in Computer Science and Engineering at POSTECH, working in the Software Verification Lab under the supervision of Prof. Kyungmin Bae. My research focuses on applying formal methods to improve safety and reliability of safety-critical systems, specifically neural networks and Trusted Execution Environment (TEE).
Research Interests
- Formal Verification of Neural Networks
- Incremental verification: scalable techniques for verifying optimized neural networks
- Formal Specification and Verification of TEE Components
- Formal specification: modeling TEE APIs and TEE applications
- Formal verification: model checking of TEE applications
Education
- Ph.D in Computer Science and Engineering, POSTECH (Advisor: Prof. Kyungmin Bae)
- 09.2021 ~ (Pursuing)
- B.S in Computer Science and Engineering, POSTECH
- 02.2018 ~ 08.2021
Publications
G. Yu, S. Chae, K. Bae, and S. Moon, Formal Specification of Trusted Execution Environment APIs, International Conference on Fundamental Approaches to Software Engineering (FASE), 2024 [paper] [github]
J. Yeon, S. Chae, and K. Bae, Layered Abstraction Technique for Effective Formal Verification of Deep Neural Networks, Journal of KIISE, Vol. 49, No. 11, Nov 2022 (in Korean) (Excellent paper award) [paper]
J. Yeon, S. Chae, and K. Bae, Layered Abstraction for Formally Verifying Deep Neural Networks, Korea Software Congress (KSC), Dec 20-22, 2021 (in Korean) (Best paper award) [paper]
Awards
- ETAPS 2024 Student Scholarship, 02.2024 (ETAPS e.V. and University of Luxembourg)
- Bang Seung-yang Graduate Fellowship, 2022 (Department of CSE, POSTECH)
- Global Leadership Program (GLP), Sophomore-Junior, Junior-Senior (Department of CSE, POSTECH)
Teaching Experiences
- (CSED490H) Programming Studio, Teaching Assistant (Spring 2024)
- (CSED504) Advanced Operating System, Teaching Assistant (Spring 2023)
- (CSED420) Software Verification, Teaching Assistant (Spring 2022)
Talks
- Utilization of Activation Property DB for Efficient Incremental Neural Network Verification, 9th Summer STAAR Workshop, Jul 27, 2025 (in Korean) [slides]