UMBC Center for AI ]]>
Joint work with Yuechun Gu and Jiajie He
To appear, 2015 Int. Conf. on Knowledge Discovery and Data Mining
With increasingly deployed deep neural networks in sensitive application domains, such as healthcare and security, it is essential to understand what kind of sensitive information can be inferred from these models. Most known model-targeted attacks assume attackers have learned the application domain or training data distribution to ensure successful attacks. Can removing the domain information from model APIs protect models from these attacks? Our work studies this critical problem. Unfortunately, even with minimal knowledge, i.e., accessing the model as an unnamed function without leaking the meaning of input and output, the proposed adaptive domain inference (ADI) attack can still successfully estimate relevant subsets of training data. We show that the extracted relevant data can significantly improve the performance of model-inversion attacks, for instance. Specifically, the ADI method uses the concept hierarchy extracted from the public and private datasets that the attacker can access, and it applies a novel algorithm to adaptively tune the likelihood of leaf concepts in the hierarchy showing up in the unseen training data. For comparison, we also designed a straightforward hypothesis-testing-based attack called LDI. Among all candidate methods, the ADI attack extracts partial training data at the concept level, converges fastest, and requires the fewest target-model accesses.
Dr. Keke Chen is an associate professor in the UMBC CSEE Department. His recent research focuses on privacy and security issues with AI model training and deployment. He earned his PhD in computer science from Georgia Tech in 2006. Before joining UMBC, he was a Northwestern Mutual associate professor of computer science at Marquette University.
Support for this event was provided in part by the NSF under SFS grant DGE-1753681.
UMBC professor Mohammad Samarah will give a practical webinar that breaks down key concepts in artificial intelligence and explores how you can leverage them in your personal and professional life. Whether you’re curious or cautious about AI, this session will help you engage with it more confidently and strategically. Dr. Samarah is the graduate program director for UMBC’s Masters of Professional Studies in Software Engineering program.
Get more information and join the webinar at Noon on April 16 here.
The webinar is part of the Paws & Pivot webinar series that provides a platform for recently laid-off federal workers and other affected professionals to connect, learn, and thrive. Its expert-led sessions focus on providing practical resources and fostering a community of practice centered on successful career transitions, essential upskilling opportunities, and leveraging available community support. ]]>
UMBC Campus Life will hold workshops on Ethical Leadership in the Age of AI on Tuesday, April 8, 2025, in Commons 331. The workshop is aimed at students, but faculty and staff are welcome to attend as well. To accommodate schedules, the 90-minute workshop will be held starting at 10 am, Noon, and 2 pm.
The workshop will address what it means to engage with AI systems in an increasingly digital world and cover the principles of ethical leadership, AI challenges and governance, ethical dilemmas with AI, and more.
