Speaker:
Shagufta Mehnaz
The Pennsylvania State University

Abstract: The increased use of machine learning (ML) technologies on proprietary and sensitive datasets has led to increased privacy breaches in many sectors, including healthcare and personalized medicine. Although federated learning (FL) systems allow multiple parties to train ML models collaboratively without sharing their raw data with third-party entities, security concerns arise from the involvement of potentially malicious FL clients aiming to disrupt the learning process. In this talk, I will present how my research addresses these challenges by developing frameworks to analyze and improve the privacy and security aspects of ML. First, I will talk about model inversion attacks that allow an adversary to infer part of the sensitive training data with only black-box access to a vulnerable classification model. I will then present FLShield, a novel FL framework that utilizes benign data from FL participants to validate the local models before taking them into account for generating the global model. I will conclude with a discussion of challenges in building practical data-driven systems that take into account data privacy and security while keeping the intended functionality of the system unimpaired.

About: Shagufta Mehnaz is an Assistant Professor of the Computer Science and Engineering department at The Pennsylvania State University. She is broadly interested in the areas of privacy, security, and machine learning. Her research focuses on enhancing the privacy and security of machine learning techniques and models themselves, as well as developing novel machine learning techniques to protect data security and privacy. She directs the PRIvacy, Security, and Machine Learning lab (PRISMLab) at Penn State. She obtained her Ph.D. in Computer Science from Purdue University in 2020. She also received the Bilsland Dissertation Fellowship at Purdue. She was one of the 100 Computer Science Young Researchers selected worldwide for the Heidelberg Laureate Forum (HLF) in 2018.

The weekly security seminar has been held every semester since spring of 1992. We invite personnel at Purdue and visitors from outside to present on topics of particular interest to them in the areas of computer and network security, computer crime investigation, information warfare, information ethics, public policy for computing and security, the computing "underground," and other related topics. More info