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

With the wide application of deep neural networks, it is important to verifya host's possession over a deep neural network model and protect the model. Tomeet this goal, various mechanisms have been designed. By embedding extrainformation into a network and revealing it afterward, the watermark becomes acompetitive candidate in proving integrity for deep learning systems. However,concurrent watermarking schemes can hardly be adopted for emerging distributedlearning paradigms that raise extra requirements during the ownershipverification. A spearheading distributed learning paradigm is federatedlearning (FL) where many parties participate in training one single model. Eachauthor participating in the FL should be able to verify its ownershipindependently. Moreover, there are other potential threat and correspondingsecurity requirements under this scenario. To meet those requirements, in thispaper, we demonstrate a watermarking protocol for protecting deep neuralnetworks in the setting of FL. By incorporating the state-of-the-artwatermarking scheme and the cryptological primitive designed for distributedstorage, the protocol meets the need for ownership verification in the FLscenario without violating the privacy for each participant. This work pavesthe way for generalizing watermark as a practical security mechanism forprotecting deep learning models in distributed learning platforms.

Towards Practical Watermark for Deep Neural Networks in Federated Learning