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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Training high-performing deep learning models require a rich amount of datawhich is usually distributed among multiple data sources in practice. Simplycentralizing these multi-sourced data for training would raise criticalsecurity and privacy concerns, and might be prohibited given the increasinglystrict data regulations. To resolve the tension between privacy and datautilization in distributed learning, a machine learning framework calledprivate aggregation of teacher ensembles(PATE) has been recently proposed. PATEharnesses the knowledge (label predictions for an unlabeled dataset) fromdistributed teacher models to train a student model, obviating access todistributed datasets. Despite being enticing, PATE does not offer protectionfor the individual label predictions from teacher models, which still entailsprivacy risks. In this paper, we propose SEDML, a new protocol which allows tosecurely and efficiently harness the distributed knowledge in machine learning.SEDML builds on lightweight cryptography and provides strong protection for theindividual label predictions, as well as differential privacy guarantees on theaggregation results. Extensive evaluations show that while providing privacyprotection, SEDML preserves the accuracy as in the plaintext baseline.Meanwhile, SEDML's performance in computing and communication is 43 times and1.23 times higher than the latest technology, respectively.

SEDML: Securely and Efficiently Harnessing Distributed Knowledge in Machine Learning