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

Federated learning (FL) emerges as an effective collaborative learningframework to coordinate data and computation resources from massive anddistributed clients in training. Such collaboration results in non-trivialintellectual property (IP) represented by the model parameters that should beprotected and shared by the whole party rather than an individual user.Meanwhile, the distributed nature of FL endorses a malicious client theconvenience to compromise IP through illegal model leakage to unauthorizedthird parties. To block such IP leakage, it is essential to make the IPidentifiable in the shared model and locate the anonymous infringer who firstleaks it. The collective challenges call for \emph{accountable federatedlearning}, which requires verifiable ownership of the model and is capable ofrevealing the infringer's identity upon leakage. In this paper, we proposeDecodable Unique Watermarking (DUW) for complying with the requirements ofaccountable FL. Specifically, before a global model is sent to a client in anFL round, DUW encodes a client-unique key into the model by leveraging abackdoor-based watermark injection. To identify the infringer of a leakedmodel, DUW examines the model and checks if the triggers can be decoded as thecorresponding keys. Extensive empirical results show that DUW is highlyeffective and robust, achieving over $99\%$ watermark success rate for Digits,CIFAR-10, and CIFAR-100 datasets under heterogeneous FL settings, andidentifying the IP infringer with $100\%$ accuracy even after common watermarkremoval attempts.

Who Leaked the Model? Tracking IP Infringers in Accountable Federated Learning