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

Since there are multiple parties in collaborative learning, malicious partiesmight manipulate the learning process for their own purposes through backdoorattacks. However, most of existing works only consider the federated learningscenario where data are partitioned by samples. The feature-partitionedlearning can be another important scenario since in many real worldapplications, features are often distributed across different parties. Attacksand defenses in such scenario are especially challenging when the attackershave no labels and the defenders are not able to access the data and modelparameters of other participants. In this paper, we show that even parties withno access to labels can successfully inject backdoor attacks, achieving highaccuracy on both main and backdoor tasks. Next, we introduce several defensetechniques, demonstrating that the backdoor can be successfully blocked by acombination of these techniques without hurting main task accuracy. To the bestof our knowledge, this is the first systematical study to deal with backdoorattacks in the feature-partitioned collaborative learning framework.

Backdoor attacks and defenses in feature-partitioned collaborative learning