Fair and Secure Federated Learning for Generative AI

Generative artificial intelligence Gen AI has evolved rapidly, but is increasingly limited by data availability and privacy concerns. Federated Learning FL, including federated fine tuning, offers a promising, decentralised approach to collaboratively train models without sharing raw data. By preserving privacy and integrating more diverse data, FL can enhance generalisation. However, uneven data distribution risks skewing model fairness, and the open nature of FL makes it vulnerable to data poisoning attacks. This PhD project aims to develop robust defense mechanisms that ensure fairness, security, and reliable performance in federated fine tuning for real-world Gen AI applications.

This project is supported by the Research project scholarship.

Learn more about the Research project scholarship.

Your application will be assessed on a competitive basis.

We take into account your:

• previous academic record
• publication record
• honours and awards
• employment history

A working knowledge of advanced machine learning frameworks (e.g., TensorFlow or PyTorch), GPU-based computing, and distributed systems would be of benefit to someone working on this project.

You’ll demonstrate academic achievement in the fields of computer science, data science, cybersecurity or related disciplines and show potential for scholastic success.

A background or knowledge of distributed systems, federated learning, adversarial machine learning, and generative AI is highly desirable.

To be considered for this scholarship, please email the following documents to Dr Azadeh Ghari-Neiat (a.gharineiat@uq.edu.au):

• Cover letter
• CV
• Academic transcript/s
• Evidence for meeting UQ's English language proficiency requirements eg TOEFL, IELTS

Please note the following: Submitting the above documents does not constitute a full application for admission into The University of Queensland's PhD program. If you are selected as the preferred applicant, you will then be invited to submit a full application for admission. You can familiarise yourself with the documents required for this process on the UQ Study website.