"LSU Computer Science Professor Leading Project to Increase Security in Federated Learning"
Federated learning has garnered attention for its potential to bolster privacy, security, and efficacy across multiple industries. This technique is sometimes subjected to "critical learning" to improve its quality and robustness. However, during these times, external actors have the opportunity to initiate precise and damaging attacks. To better understand these opportunities and attacks, Louisiana State University Computer Science Assistant Professor Hao Wang is collaborating with Assistant Professor Jian Li, from the Department of Computer Science at Stony Brook University, and Associate Professor Xu Yuan, from the Department of Computer and Information Sciences at the University of Delaware. Their work aims to deliver a prototype federated learning system with algorithms that detect critical learning periods and use attack/defense methods. According to Wang, a critical learning period is an inherent property of the training process of deep learning models. He adds that it could boost various attacks, including data-poisoning attacks and model-poisoning attacks. This article continues to discuss the project aimed at improving security in federated learning.