Efficient Secure Computation

Secure computation allows for computation of arbitrary functions directly on encrypted data and hides all information about the data against untrusted parties, even if the untrusted parties are involved in the computation. Secure computation could enable, for example, companies to outsource and process their data in untrusted clouds, or two mutually untrusted parties to mine their datasets together. However, secure computation currently is often too inefficient to be practical. Despite recent huge improvements, secure computation is still tens of thousand to billions times slower than computation in the clear. This becomes a major impediment to widespread use of secure computation.

I started my research on efficient secure computation in 2011. I tackle this problem from complete new angles, by designing cryptographic data structures and associated protocols that allow more efficient and scalable secure computation and by building efficient secure computation protocols using recently developed cryptographic primitives. The protocols have been motivated by and applied in solving security/private issues in cloud computing, data mining, and machine learning.

Related publications

  1. FC
    Multi-party Updatable Delegated Private Set Intersection
    Aydin Abadi,  Changyu Dong, Steven Murdoch, and Sotirios Terzis
    In 26th International Conference on Financial Cryptography and Data Security, 2022
  2. USENIX
    How to Make Private Distributed Cardinality Estimation Practical, and Get Differential Privacy for Free
    Changhui Hu, Jin Li, Zheli Liu, Xiaojie Guo, Yu Wei, Xuan Guang, Grigorios Loukides, and Changyu Dong
    In 30th USENIX Security Symposium, 2021
  3. TKDE
    MAS-Encryption and Its Applications in Privacy-Preserving Classifiers
    Chongzhi Gao, Jin Li, Shibing Xia, Kim-Kwang Raymond Choo, Wenjing Lou, and Changyu Dong
    IEEE Trans. Knowl. Data Eng., 2020
  4. TDSC
    Efficient Delegated Private Set Intersection on Outsourced Private Datasets
    Aydin Abadi, Sotirios Terzis, Roberto Metere, and Changyu Dong
    IEEE Trans. Dependable Secur. Comput., 2019
  5. TIFS
    Approximating Private Set Union/Intersection Cardinality With Logarithmic Complexity
    Changyu Dong, and Grigorios Loukides
    IEEE Trans. Inf. Forensics Secur., 2017
  6. FC
    VD-PSI: Verifiable Delegated Private Set Intersection on Outsourced Private Datasets
    Aydin Abadi, Sotirios Terzis, and Changyu Dong
    In 20th International Conference on Financial Cryptography and Data Security, 2016
  7. SEC
    O-PSI: Delegated Private Set Intersection on Outsourced Datasets
    Aydin Abadi, Sotirios Terzis, and Changyu Dong
    In 30th IFIP TC 11 International Conference on ICT Systems Security and Privacy Protection, 2015
  8. SPW
    Efficient Data Intensive Secure Computation: Fictional or Real?
    Changyu Dong
    In 23rd International Workshop on Security Protocols, 2015
  9. ESORICS
    A Fast Single Server Private Information Retrieval Protocol with Low Communication Cost
    Changyu Dong, and Liqun Chen
    In 19th European Symposium on Research in Computer Security, 2014
  10. PAKDD
    A Fast Secure Dot Product Protocol with Application to Privacy Preserving Association Rule Mining
    Changyu Dong, and Liqun Chen
    In Pacific-Asia Conference on Knowledge Discovery and Data Mining, 2014
  11. SAC
    Efficient protocols for private record linkage
    Zikai Wen, and Changyu Dong
    In ACM Symposium on Applied Computing, 2014
  12. CCS
    When private set intersection meets big data: an efficient and scalable protocol
    Changyu Dong, Liqun Chen, and Zikai Wen
    In ACM SIGSAC Conference on Computer and Communications Security, 2013
  13. DBSec
    Fair Private Set Intersection with a Semi-trusted Arbiter
    Changyu Dong, Liqun Chen, Jan Camenisch, and Giovanni Russello
    In 27th Annual IFIP WG 11.3 Conference on Data and Applications Security and Privacy, 2013