홍상현 오레곤 주립 대학교

Sanghyun Hong

Assistant Professor
Computer Science
Oregon State University

       

Computer Security, Privacy, and Machine Learning

I am an Assistant Professor of Computer Science at Oregon State University, working at the intersection of computer security, privacy, and machine learning.

Research Interests

My research objective is to build secure and reliable machine learning systems from a systems security perspective. To this end, I focus on characterizing the computational properties of deep neural networks which makes the networks particularly vulnerable in practical adversarial settings.

If you want to know more about my work, I recommend watching my talk at USENIX Enigma 2021 (TED Talk for Security!) or reading my blog posts or my work featured in MIT Technology Review, TechTalks, and DEV Podcast.

Bio

2021, Ph.D. in Computer Science, University of Maryland, College Park
2017, M.S. in Computer Science, University of Maryland, College Park
2015, B.S. in Electrical Engineering and Computer Science, Seoul Nat'l University


I am actively looking for talented, self-motivated students. Please fill out this form if you're interested.

Contacts

 Email: sanghyun.hong [at] oregonstate [dot] edu
 Office: 4103 Kelley Engineering Center (KEC), 2500 NW Monroe Ave, Corvallis, OR 97331 USA
 Office Hours: Tu / Th: 2:00 - 3:00 pm PST (only for OSU students, email me for an appointment)


Announcements

Sep. 28, 2021
My Qu-ANTI-zation paper is accepted at NeurIPs 2021! [New]
Sep. 16, 2021
I joined CS @ Oregon State University as an Assistant Professor. [New]
Jul. 23, 2021
I successfully defended my dissertation.
Jan. 14, 2021
My DeepSloth paper accepted at ICLR as a spotlight!
Oct. 4, 2020
I started my internship at Google Brain under the supervision of Dr. Nicholas Carlini and Dr. Alexey Kurakin.
Oct. 1, 2020
My talk proposal is accepted to USENIX Enigma 2021!
May. 26, 2020
I became a PhD candidate :)

Selected Publications [Full list, Google Scholar]

Qu-ANTI-zation: Exploiting Neural Network Quantization for Achieving Adversarial Outcomes
Sanghyun Hong, Michael-Andrei Panaitescu-Liess, Yigitcan Kaya, and Tudor Dumitraș
ICLR 21
[Spotlight]
A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference
*Sanghyun Hong, *Yigitcan Kaya, Ionuţ-Vlad Modoranu, and Tudor Dumitraș
(* indicates equal contribution)
How to 0wn NAS in Your Spare Time
Sanghyun Hong, Michael Davinroy, Yigitcan Kaya, Dana Dachman-Soled, and Tudor Dumitraș
Terminal Brain Damage: Exposing the Graceless Degradation in Deep Neural Networks Under Hardware Fault Attacks
Sanghyun Hong, Pietro Frigo, Yigitcan Kaya, Cristiano Giuffrida, and Tudor Dumitraș
Shallow-Deep Networks: Understanding and Mitigating Network Overthinking
Yigitcan Kaya, Sanghyun Hong, and Tudor Dumitraș
Go Serverless: Securing Cloud via Serverless Design Patterns
Sanghyun Hong, Abhinav Srivastava, William Shambrook, and Tudor Dumitraș

Students

Graduate
Hoang Le (PhD, Computer Science; Winter 2022 - Present)
Committee
Jarrod Jeffrey Isaac Hollis (PhD, Computer Science; Fall 2021 - Present)

Teaching

Spring 2022
CS 344: Operating Systems I
Winter 2022
CS 499/599: Special Topics: Machine Learning Security [Course website]

Services

Program Comittee
2021: ACM CCS 2021 Workshop on Artificial Intelligence and Security (AISec)
2021: ACM CCS 2021 Workshop on Privacy in the Electronic Society (WPES)
2021: International Symposium on Research in Attacks, Intrusions and Defenses (RAID)
2021: IEEE S&P 2021 Workshop on Deep Learning Security (DLS)
2020: ICLR 2020 Workshop on Towards Trustworthy ML: Rethinking Security and Privacy for ML
Reviewer
2022, 2021: International Conference on Learning Representations (ICLR)
2021, 2020: Conference on Neural Information Processing Systems (NeurIPs)
2021, 2020: International Conference on Machine Learning (ICML) [Top 33% Reviewer]
2021: IEEE Access
2019: Computer & Security
2018: IEEE Transaction on Cloud Computing (TCC)
Sub-reviewer
2020, 2019, 2017: Network and Distributed System Security Symposium (NDSS)
2019, 2017: IEEE Symposium on Security and Privacy (IEEE S&P; Oakland)
2019, 2018, 2017: ACM Symposium on Computer and Communications Security (CCS)
2018, 2017, 2016: USENIX Security Symposium (USENIX)
2019, 2018: International Symposium on Research in Attacks, Intrusions and Defenses (RAID)