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

Split learning enables efficient and privacy-aware training of a deep neuralnetwork by splitting a neural network so that the clients (data holders)compute the first layers and only share the intermediate output with thecentral compute-heavy server. This paradigm introduces a new attack medium inwhich the server has full control over what the client models learn, which hasalready been exploited to infer the private data of clients and to implementbackdoors in the client models. Although previous work has shown that clientscan successfully detect such training-hijacking attacks, the proposed methodsrely on heuristics, require tuning of many hyperparameters, and do not fullyutilize the clients' capabilities. In this work, we show that given modestassumptions regarding the clients' compute capabilities, an out-of-the-boxoutlier detection method can be used to detect existing training-hijackingattacks with almost-zero false positive rates. We conclude through experimentson different tasks that the simplicity of our approach we name SplitOut makesit a more viable and reliable alternative compared to the earlier detectionmethods.

SplitOut: Out-of-the-Box Training-Hijacking Detection in Split Learning via Outlier Detection