English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Secure multi-party computation-based machine learning, referred to as MPL,has become an important technology to utilize data from multiple parties withprivacy preservation. While MPL provides rigorous security guarantees for thecomputation process, the models trained by MPL are still vulnerable to attacksthat solely depend on access to the models. Differential privacy could help todefend against such attacks. However, the accuracy loss brought by differentialprivacy and the huge communication overhead of secure multi-party computationprotocols make it highly challenging to balance the 3-way trade-off betweenprivacy, efficiency, and accuracy.  In this paper, we are motivated to resolve the above issue by proposing asolution, referred to as PEA (Private, Efficient, Accurate), which consists ofa secure DPSGD protocol and two optimization methods. First, we propose asecure DPSGD protocol to enforce DPSGD in secret sharing-based MPL frameworks.Second, to reduce the accuracy loss led by differential privacy noise and thehuge communication overhead of MPL, we propose two optimization methods for thetraining process of MPL: (1) the data-independent feature extraction method,which aims to simplify the trained model structure; (2) the local data-basedglobal model initialization method, which aims to speed up the convergence ofthe model training. We implement PEA in two open-source MPL frameworks:TF-Encrypted and Queqiao. The experimental results on various datasetsdemonstrate the efficiency and effectiveness of PEA. E.g. when ${\epsilon}$ =2, we can train a differentially private classification model with an accuracyof 88% for CIFAR-10 within 7 minutes under the LAN setting. This resultsignificantly outperforms the one from CryptGPU, one SOTA MPL framework: itcosts more than 16 hours to train a non-private deep neural network model onCIFAR-10 with the same accuracy.

Private, Efficient, and Accurate: Protecting Models Trained by Multi-party Learning with Differential Privacy