
FedPLVM introduces dual-level prototype clustering and an α-sparsity prototype loss to mitigate cross-domain feature representation variance in federated learning.


coremsg

addressing-cross-domain-variance-in-federated-prototype-learning-with-heterogeneous-data-domains


Addressing Cross-Domain Variance in Federated Prototype Learning with Heterogeneous Data Domains


title_rewrite


Dual-level prototype clustering and α-sparsity loss mitigate cross-domain variance in Federated Prototype Learning.


fedplvm-addressing-cross-domain-variance-in-federated-prototype-learning-with-heterogeneous-data-domains


FedPLVM: Addressing Cross-Domain Variance in Federated Prototype Learning with Heterogeneous Data Domains
