user_image

Tu Hoang

I am a senior researcher in privacy, security, machine learning, and decentralized systems. I have presented privacy-preserving technologies for machine learning/deep learning on various data types (i.e., knowledge graphs, relational data, textual data) and decentralized systems (i.e., Hyperledger Fabric). The techniques require me to have strong knowledge of state-of-the-art machine learning/deep learning models to analyze their vulnerabilities and protect them. Moreover, I am a reviewer of high-quality journals and conferences (e.g., VLDB, TOPS).

EXPERIENCE

  • 2020-12-01 – 2022-11-30

    University of Insubria (Varese, Italy), Research Fellow

  • 2017-10-01 – 2020-12-01

    University of Insubria (Varese, Italy), PhD Student

  • 2013-03-24 – 2013-09-18

    National Institute of Informatics (Tokyo, Japan), Research Intern

  • 2009-09-01 – 2017-10-01

    University of Science, Vietnam National University (Ho Chi Minh City, Vietnam), Lecturer & Teaching Assistant

AWARDS

  • 1 of the 5 best creative projects in Microsoft's Imagine Cup 2009 competition in Vietnam

RESEARCH INTERESTS

  • Privacy: Differential Privacy k-Anonymity
  • Machine Learning: Federated Learning Knowledge Graph Embedding Node2Vec Image Classification
  • Decentralized Systems: Blockchain NoSQL
  • Cryptography: Multi-party Computation Homomorphic Encryption Pedersen Commitment Zero-Knowledge-Proof
  • Algorithm Design: Graph Theory Discrete Optimization

SKILLS

  • Programming: Python Go Node.js Solidity Bash Scala Ruby(Rails) Swift C#
  • Machine Learning: PyTorch Tensorflow MXNet Scikit-Learn Numpy Pandas Matplotlib
  • Blockchain: Hyperledger Fabric Ethereum
  • Tools: Docker Git Linux Microsoft Azure

RECENT PROJECTS

2019

Directed Graph Anonymization
#Python #Algorithm Design #Discrete Optimization
An anonymization algorithm for static directed graphs.

2019

Static Knowledge Graph Anonymization
#Python #PyTorch #Algorithm Design #Discrete Optimization #Scikit-Learn
A k-anonymity approach for publishing static knowledge graphs.

2020

Sequential Knowledge Graph Anonymization
#Python #Algorithm Design #Discrete Optimization #Scikit-Learn
A k-anonymity approach for sequentially publishing knowledge graphs.

2021

Personalized Knowledge Graph Anonymization
#Python #Algorithm Design #Discrete Optimization #Scikit-Learn
A k-anonymity approach allowing data owners to specify their anonymity levels.

2022

Preventing Attribute-Leakage in Sequential Publishing of Knowledge Graphs
#Python #Algorithm Design #Discrete Optimization #Scikit-Learn
Preventing attribute-leakage attack and identity-leakage while allowing data providers to freely modify their knowledge graphs.

2022 – Developing

Blockchain-Based Anonymization
#Go #Hyperledger Fabric #Node.js #Homomorphic Encryption #Zero-Knowledge-Proof #Docker
A transparent platform for data owners, data providers, and data recipients to collect, publish anonymized knowledge graphs, and verify whether anonymized knowledge graphs satisfy k-anonymity.

2021 – Developing

Promark: A Transparent Blockchain-based Marketing Platform
#Go #Hyperledger Fabric #Node.js #Homomorphic Commitment #Multi-Party Computation #Zero-Knowledge-Proof #Docker
A transparent platform for publishers, marketers, and customers to distribute marketing information and evaluate the effectiveness of marketing campaigns.

RECENT PUBLICATIONS

[All Publications]
  1. Privacy-preserving Decentralized Learning of Knowledge Graph Embeddings
    Hoang, Anh-Tu, Lekssays, Ahmed, Carminati, Barbara, and Ferrari, Elena
    In Proceedings of the Workshops of the EDBT/ICDT 2023 Joint Conference, Ioannina, Greece, March, 28, 2023, vol. 3379, 2023
  2. Time-Aware Anonymization of Knowledge Graphs (Accepted)
    Hoang, Anh-Tu, Carminati, Barbara, and Ferrari, Elena
    In the ACM Transactions on Privacy and Security (TOPS), vol. , pp. , 2022
  3. Privacy-Preserving Sequential Publishing of Knowledge Graphs
    Hoang, Anh-Tu, Carminati, Barbara, and Ferrari, Elena
    In 2021 IEEE 37th International Conference on Data Engineering (ICDE), vol. , pp. 2021-2026, 2021