Chau-Wai Wong
North Carolina State University, NC·Engineering·Early-career
Sharing data across many computers or organizations can make artificial intelligence (AI) smarter, but it can also leak private information or let fakes slip through. This work tackles two security problems where data, images and trust meet.
- Fields
Engineering
Computer Science
- Topics
federated learning
data privacy
data reconstruction attacks
- Grants
- 2 known grants, through Sep 2029
What they work on
- Develop a privacy-leakage metric for images rebuilt by reconstruction attacks in federated learning that matches human judgments of privacy, using user studies to design and test it.
- Analyze existing defenses against honest-but-curious clients, participants who covertly extract others' data in federated learning, and develop mathematically grounded defenses that close their gaps.
- Design efficient, scalable computer vision algorithms that identify individual products from physically unclonable features (PUFs), random microscopic surface textures, imaged with consumer-grade mobile cameras.
- Build models of how light reflects from microscopic surface structures, combining physics-guided signal analysis and machine learning with ambient light estimation, to extract reliable surface fingerprints.
- Integrate surface PUF verification with a blockchain-based supply-chain information system and use information-theoretic modeling to find bottlenecks in verification performance.
More about the research
One project protects the privacy of people whose data trains AI models collaboratively. The other uses tiny surface patterns on physical products, captured with ordinary phone cameras, to prove that goods in a supply chain are genuine.
Why it matters
Federated learning is spreading in healthcare, finance and public services, where leaked patient or customer data can cause real harm. Better privacy measures and defenses could make this kind of data sharing safer. Separately, phone-based product fingerprinting could help stop old electronic parts from being resold as new and keep counterfeit medicines from reaching patients, protecting public health and national security.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Joann Chen
San Diego State University, CA·Computer Science
Identify privacy risks in provenance-based machine learning anomaly detectors by applying property inference, membership inference and graph reconstruction attacks to them.
Joseph Camp
Southern Methodist University, TX·Computer Science
Develop an AI curriculum covering federated learning, explainable AI, homomorphic encryption and neural networks, taught through lectures, exercises and simulations and shared via the NAIRR Pilot.
Vanessa Chen
Carnegie Mellon University, PA·Engineering·Early-career
Design bio-inspired, time-domain analog classifiers that exploit transistor mismatch and nonlinearity to perform energy-efficient hierarchical classification at low supply voltages within a sensor.
The school
North Carolina State University: 40 labs on PI Pages, $717M a year on research, 603 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.