Hello! I am a researcher at DeepingSource specializing in privacy-preserving algorithms for AI. My focus is on developing innovative solutions that protect sensitive data while enabling their effective and ethical use, in addition to identifying novel privacy risks in AI applications.
<aside> π₯ Korea University B.S. in Computer Science and Engineering March 2017 - August 2021
</aside>
<aside> π₯ Korea Advanced Institute of Science and Technology (KAIST) M.S. student in Graduate School of Information Security September 2021 - February 2023
</aside>
Learning to Generate Inversion-Resistant Model Explanations Hoyong Jeong, Suyoung Lee, Sung Ju Hwang, Sooel Son *The 36th Conference on Neural Information Processing Systems (NeurIPS), 2022 *****[paper], [artifact]
Exploiting Metaobjects to Reinforce Data Leakage Attacks Hoyong Jeong, Hodong Kim, Junbeom Hur The 25th International Symposium on Research in Attacks, Intrusions and Defenses (RAID), 2022 [paper]
μ΄λ―Έμ§ μΈμ κΈ°λ°μ ν΅ν©μ FPS κ²μ μν°μΉνΈ μ루μ μ νΈμ©, μ΄μμ, μ΄μ°¬μ°, μ μ§μ©, κΉλν, μ΄κ²½ν, μ¬μμ§, μ μν μ 보보νΈνν λκ³νμ λν (CISC-W), 2021 [paper] **
Neural Network Stealing via Meltdown Hoyong Jeong, Dohyun Ryu, Junbeom Hur The 35th International Conference on Information Networking (ICOIN), 2020 [paper]
Web Security & Privacy Laboratory, KAIST (advisor: Prof. Sooel Son) Masterβs Student (2021.09 - 2023.02)
- Analyzed privacy risks in ML applications
- Developed an obfuscation algorithm for private release of XAI features
Information System Security Laboratory, Korea University (advisor: Prof. Junbeom Hur) Undergraduate Research Intern (2019.03 β 2021.08)
- Analyzed microarchitectural vulnerabilities of the CPU
- Developed neural network stealing algorithm leveraging microarchitectural vulnerabilities