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

Federated learning (FL) is a heavily promoted approach for training ML modelson sensitive data, e.g., text typed by users on their smartphones. FL isexpressly designed for training on data that are unbalanced and non-iid acrossthe participants. To ensure privacy and integrity of the fedeated model, latestFL approaches use differential privacy or robust aggregation.  We look at FL from the \emph{local} viewpoint of an individual participantand ask: (1) do participants have an incentive to participate in FL? (2) howcan participants \emph{individually} improve the quality of their local models,without re-designing the FL framework and/or involving other participants?  First, we show that on standard tasks such as next-word prediction, manyparticipants gain no benefit from FL because the federated model is lessaccurate on their data than the models they can train locally on their own.Second, we show that differential privacy and robust aggregation make thisproblem worse by further destroying the accuracy of the federated model formany participants.  Then, we evaluate three techniques for local adaptation of federated models:fine-tuning, multi-task learning, and knowledge distillation. We analyze whereeach is applicable and demonstrate that all participants benefit from localadaptation. Participants whose local models are poor obtain big accuracyimprovements over conventional FL. Participants whose local models are betterthan the federated model\textemdash and who have no incentive to participate inFL today\textemdash improve less, but sufficiently to make the adaptedfederated model better than their local models.

Salvaging Federated Learning by Local Adaptation