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 has shown great potentials for the distributed datautilization and privacy protection. Most existing federated learning approachesfocus on the supervised setting, which means all the data stored in each clienthas labels. However, in real-world applications, the client data are impossibleto be fully labeled. Thus, how to exploit the unlabeled data should be a newchallenge for federated learning. Although a few studies are attempting toovercome this challenge, they may suffer from information leakage or misleadinginformation usage problems. To tackle these issues, in this paper, we propose anovel federated semi-supervised learning method named FedTriNet, which consistsof two learning phases. In the first phase, we pre-train FedTriNet usinglabeled data with FedAvg. In the second phase, we aim to make most of theunlabeled data to help model learning. In particular, we propose to use threenetworks and a dynamic quality control mechanism to generate high-qualitypseudo labels for unlabeled data, which are added to the training set. Finally,FedTriNet uses the new training set to retrain the model. Experimental resultson three publicly available datasets show that the proposed FedTriNetoutperforms state-of-the-art baselines under both IID and Non-IID settings.

FedTriNet: A Pseudo Labeling Method with Three Players for Federated Semi-supervised Learning