AFCL solves federated continual learning without gradients by aggregating local Gram matrices and label statistics, proving exact spatio-temporal invariance: the global model equals centralized joint learning for any data partition.
Ten challenging problems in federated foundation models, 2025
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
AFCL: Analytic Federated Continual Learning for Spatio-Temporal Invariance of Non-IID Data
AFCL solves federated continual learning without gradients by aggregating local Gram matrices and label statistics, proving exact spatio-temporal invariance: the global model equals centralized joint learning for any data partition.