Life is about exploring
From 2021 Nov., I am a Posdoc at Tsinghua University, working with Prof. Xiaolin Hu. From 2018 to 2021, I was PhD fellow in Audio Analysis Lab at Aalborg University, Denmark. I worked on multi-discipline including distributed signal processing, secure multiparty computation, privacy and security. My PhD thesis is 'Privacy-Preserving distributed processing over networks' under supervision of Professor Mads Græsbøll Christensen and Professor Richard Heusdens. In 2019, I visited CAS group in Delft University of Technology (TUDelft) as a visiting PhD.
Email: qiongxiuli[at]outlook.com, qiongxiuli[at]fudan.edu.cn
Involved project
Sensing the world without violating privacy
Driven by the concern of private data abuse in the current 'Big Data' word. I was working on a very interesting project called 'SECURE' (Secure Estimation and Control Using Recursion and Encryption) at Aalborg University, Denmark. It is an interdisciplinary project that requires techniques from distributed optimization, information theory, statistics, differential privacy, cryptography, security, and more. My main research focus is to design privacy-preserving protocols for distributed processing over networks. I am particularly interested in computationally efficient solutions for practical scenarios such as resource-constrained wireless sensor networks.
News
[Dec. 2023]Three papers on privacy in federated learning and 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.
[May. 2022] Our paper on quantized privacy-preserving distributed average consensus is accepted in EUSIPCO 2022
[Mar. 2022] I gave a guest course lecture on 'Privacy-preserving distributed signal processing' in TUDelft for graduate students.
[Feb. 2022] Our paper on privacy-preserving distributed EM algorithm for GMM is accepted in ICASSP 2022.
[Jan. 2022] Our paper 'Communication Efficient Privacy-Preserving Distributed Optimization using Adaptive Differential Quantization' is accepted in Signal Processing.
[Aug. 2021] I successfully defended my PhD thesis entitled 'Privacy-preserving distributed processing over networks' on Aug. 30th [PDF], I am really honored to have Prof. Antonio G. Marques, Prof. Marc Moonen and Prof. Petar Popovski as my assessment committee.
[Mar. 2021] I gave a guest course lecture on 'Advanced topics in distributed signal processing' in TUDelft for graduate students.
[Jan. 2021] My presentation 'Privacy-Preserving Distributed Signal Processing 'won 3 minute thesis (3MT) contest in EUSIPCO 2020. [Link]
[Dec. 2020] Our paper "Privacy-Preserving Distributed Processing: Metrics, Bounds, and Algorithms" is accepted in IEEE Transactions on Information Forensics and Security.
[Oct. 2020] Our paper "Privacy-Preserving Distributed optimization via Subspace Perturbation: A General Framework" is accepted in IEEE Transactions on Signal Processing.
[Sep. 2020] Our paper "Privacy-Preserving Distributed Processing: Metrics, Bounds, and Algorithms" is submitted to IEEE Transactions on Information Forensics and Security.
[Aug. 2020] I gave a talk (40 minutes presentation + 20 minutes Q&A) in SPSC webinar entitled as "Privacy-Preserving Distributed optimization via Subspace Perturbation: A General Framework." 2020. Video link.
[May 2020] Two papers about privacy-preserving distributed optimization and distributed graph filtering are accepted in EUSIPCO 2020.
Publications
Journals
Qiongxiu Li, Jaron Skovsted Gundersen, Milan Lopuhaa-Zwakenberg, and Richard Heusdens. "Adaptive Differentially Quantized Subspace Perturbation (ADQSP): A Unified Framework for Privacy-Preserving Distributed Average Consensus." IEEE. Trans. on Information Forensics and Security, 2023.[Link]
Qiongxiu Li, Richard Heusdens, and Mads Græsbøll Christensen. "Communication Efficient Privacy-Preserving Distributed Optimization using Adaptive Differential Quantization." Signal Processing, 2022, [link]
Qiongxiu Li, Jaron Skovsted Gundersen, Richard Heusdens, and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Processing: Metrics, Bounds, and Algorithms." IEEE. Trans. on Information Forensics and Security, 2021, [Link]
Qiongxiu Li, Richard Heusdens, and Mads Græsbøll Christensen. "Privacy-Preserving Distributed optimization via Subspace Perturbation: A General Framework." IEEE. Trans. on Signal Processing, 2020, [link]
Conferences
Qiongxiu Li, Wenrui Yu, Changlong Ji, and Richard Heusdens. "Topology-dependent Privacy Bound for Decentralized Federated Learning", ICASSP, 2024
Qiongxiu Li, and Lixia Luo. "On the Privacy of Federated Clustering: A Cryptographic View", ICASSP, 2024
Sebastian O. Jordan，Qiongxiu Li, and Richard Heusdens. “Privacy-Preserving Distributed Optimisation Using Stochastic PDMM”, ICASSP, 2024
Qiongxiu Li, Jaron Skovsted Gundersen, Katrine Tjell, Rafal Wisniewski, and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Expectation Maximization for Gaussian Mixture model using subspace perturbation", ICASSP, 2022.
Qiongxiu Li, Milan Lopuhaa-Zwakenberg, Richard Heusdens, and Mads Græsbøll Christensen."Two for the price of one: communication efficient and privacy-preserving distributed average consensus using quantization", EUSIPCO, 2022.
Qiongxiu Li, Richard Heusdens, and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Least Squares via Subspace Perturbation." EUSIPCO, 2020.
Qiongxiu Li, Mario Coutino, Geert Leus and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Graph Filtering." EUSIPCO, 2020.
Qiongxiu Li, Richard Heusdens, and Mads Græsbøll Christensen. "Covex Optimisation-based Privacy-Preserving Distributed Average Consensus in Wireless Sensor Networks." ICASSP, 2020. [PDF]
Qiongxiu Li, Ignacio Cascudo, and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Average Consensus based on Additive Secret Sharing." EUSIPCO, 2019. [PDF]
Qiongxiu Li, and Mads Græsbøll Christensen. "A Privacy-Preserving Asynchronous Averaging Algorithm based on Shamir’s Secret Sharing." EUSIPCO, 2019. [PDF]
Presentations & Talks
Qiongxiu Li, "Privacy-Preserving Distributed Signal Processing", 3 minute thesis (3MT) winning presentation, EUSIPCO 2020.
Qiongxiu Li, Mario Coutino, Geert Leus and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Graph Filtering." EUSIPCO, 2020.
Qiongxiu Li, Richard Heusdens, and Mads Græsbøll Christensen. "Covex Optimisation-based Privacy-Preserving Distributed Least Squares via Subspace Perturbation." EUSIPCO, 2020.
Qiongxiu Li, Richard Heusdens, and Mads Græsbøll Christensen. "Covex Optimisation-based Privacy-Preserving Distributed Average Consensus in Wireless Sensor Networks." ICASSP, 2020.
Qiongxiu Li, Ignacio Cascudo, and Mads Græsbøll Christensen. "Privacy-Preserving Distributed Average Consensus based on Additive Secret Sharing." EUSIPCO, 2019.
Qiongxiu Li, and Mads Græsbøll Christensen. "A Privacy-Preserving Asynchronous Averaging Algorithm based on Shamir’s Secret Sharing." EUSIPCO, 2019.
Teaching
Teaching: Lecturer & Teaching Assistant [Fall 2018 and Fall 2019]
Program: Master of Science in Engineering; Sound and Music Computing
Aalborg University, Denmark
Distributed optimization
Teaching: Giving guest lecture on advanced topics in distributed optimization [Nov. 2018]
Lecture title: Distributed and secure computation- Applications of secure MPC protocols
Program: PhD students
Aalborg University, Denmark