Sheaf-FRL learns per-edge orthogonal or Stiefel maps to align heterogeneous agent latent spaces through a sheaf-Laplacian gluing penalty evaluated on shared pilots, with decentralized convergence guarantees.
Communication-efficient and robust multi-modal federated learning via latent-space consensus
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Sheaf-Based Federated Representation Learning
Sheaf-FRL learns per-edge orthogonal or Stiefel maps to align heterogeneous agent latent spaces through a sheaf-Laplacian gluing penalty evaluated on shared pilots, with decentralized convergence guarantees.