KIWAN
MAENG
Charles K. Etner Early Career Assistant Professor · Computer Science and Engineering · Pennsylvania State University (2022.8 --)
Postdoc Researcher · SysML Team · Meta AI Research (2021.8 -- 2022.8)
I am a tenure-track assistant professor in the
Computer Science and Engineering Department at
Pennsylvania State University. Prior to joining PSU, I did a one-year postdoc at
Meta AI Research's
SysML Team. I did my Ph.D. at
Carnegie Mellon University, under the advisement of Professor
Brandon Lucia.
My research interest lies in
systems for privacy-preserving machine learning in the broadest sense. I am interested in studying the privacy implications of modern ML systems and developing algorithm/system/hardware solutions for privacy-preserving ML training and inference. In the past, I've also worked on systems for ML, ML and sustainability, and batteryless energy-harvesting devices.
If you are interested in working with me, please send me an email
elaborating the area of research you are interested in with respect to my research interests/recent publications. Please include your relevant background experiences (e.g., courses, projects, publication experiences) as much as possible.
I am also an amateur webtoonist. Check out
my work if you know Korean. Otherwise, do not bother.
My pronouns are: he/his/him.
Email: kwmaeng91 [at] gmail [dot] com
Recent News
• Feb 2024: Our work "
Accelerating ReLU for MPC-Based Private Inference with a Communication-Efficient Sign Estimation" was accepted to
MLSys 2024!
• Dec 2023: Our work "
Compiler-based Memory Encryption for Machine Learning on Commodity Low-power Devices" was accepted to
CC 2024!
• Nov 2023: Our work "
LazyDP: Co-Designing Algorithm-Software for Scalable Training of Differentially Private Recommendation Models" was accepted to
ASPLOS 2024!
• Sep 2023: Our work "
Bounding the Invertibility of Privacy-preserving Instance Encoding using Fisher Information" was accepted to
NeurIPS 2023!
• Jul 2023: Our work "
GPU-based Private Information Retrieval for On-Device Machine Learning Inference" was accepted to
ASPLOS 2024!
• Apr 2023: Our work "
Cocktail Party Attack: Breaking Aggregation-Based Privacy in Federated Learning using Independent Component Analysis" was accepted to
ICML 2023!