Life is about exploring

Qiongxiu Li

Since 2024

Assistant Professor

Department of Electronic Systems

Copenhagen Campus

Aalborg University

Denmark

Research Focus

Signal Processing

Distributed Optimisation

Secure & Privacy Computing

Applications

Secure Optimization in AI Systems

Contacts

qiongxiuli@outlook.com

qili@es.aau.dk

Frederikskaj 12
2450 København SV
Denmark

Blogs

A New Paradigm in Federated learning via the Hidden Subset Sum Problem
Privacy through Memorization: Traditional MIAs Produce Misleading Results
An Information-Theoretical and Empirical Study: Re-Evaluating Privacy in Federated Learning (Changlong' s master project)

PhD Positions

We are recruiting multiple motivated PhD students to join the AI Defence Lab at Aalborg University (Copenhagen).

Master Thesis & Projects (6-12 months)

My projects are research-driven scientific discovery, if you're enthusiastic about diving into research and pursuing a master's project or thesis within my areas, please send me an email with your cv and research interests.

News

[Jan. 2025] Two papers on LLM and diffusion model security and alignment are accepted in NAACL2025 and ICLR2025.
[Dec. 2024] Two Papers on Distributed Machine Learning and Privacy-Preserving Non-linear Optimization have been accepted to ICASSP 2025. Congrats to Wenrui and Changlong! Preprint links: Privacy-Preserving Distributed Maximum Consensus Without Accuracy Loss and Re-Evaluating Privacy in Centralized and Decentralized Learning: An Information-Theoretical and Empirical Study
[Nov. 2024] Congratulations to Wenrui Yu on the acceptance of the paper "Provable Privacy Advantages of Decentralized Federated Learning via Distributed Optimization" in IEEE Transactions on Information Forensics and Security.
[Aug. 2024] I have given a tutorial in EUSIPCO 2024 'Privacy-Preserving Distributed Optimization: Theory, Methods and Applications' together with Prof. Richard Heusdens, Here are the slides.
[Jan. 2024] Our paper on analyzing ML privacy and robustness privacy is accepted in IEEE Transactions on Information Forensics and Security.
[Dec. 2023] Three papers on privacy in distributed optimization are accepted in ICASSP 2024.
[Dec. 2023] Our paper on privacy-preserving distributed average consensus using subspace perturbation is accepted in IEEE Transactions on Information Forensics and Security.

Page updated

Google Sites

Report abuse