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) is a promising distributed paradigm, eliminating theneed for data sharing but facing challenges from data heterogeneity.Personalized parameter generation through a hypernetwork proves effective, yetexisting methods fail to personalize local model structures. This leads toredundant parameters struggling to adapt to diverse data distributions. Toaddress these limitations, we propose FedOFA, utilizing personalized orthogonalfilter attention for parameter recalibration. The core is the Two-streamFilter-aware Attention (TFA) module, meticulously designed to extractpersonalized filter-aware attention maps, incorporating Intra-Filter Attention(IntraFa) and Inter-Filter Attention (InterFA) streams. These streams enhancerepresentation capability and explore optimal implicit structures for localmodels. Orthogonal regularization minimizes redundancy by avertinginter-correlation between filters. Furthermore, we introduce anAttention-Guided Pruning Strategy (AGPS) for communication efficiency. AGPSselectively retains crucial neurons while masking redundant ones, reducingcommunication costs without performance sacrifice. Importantly, FedOFA operateson the server side, incurring no additional computational cost on the client,making it advantageous in communication-constrained scenarios. Extensiveexperiments validate superior performance over state-of-the-art approaches,with code availability upon paper acceptance.

Energizing Federated Learning via Filter-Aware Attention