V-LETO combines evolving class prototypes with Fisher-information-based parameter freezing to handle class and feature incremental learning in vertical federated learning, reporting large gains over baselines.
J.; Das, A.; Wang, S.; and Patterson, S
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
-
Vertical Federated Continual Learning via Evolving Prototype Knowledge
V-LETO combines evolving class prototypes with Fisher-information-based parameter freezing to handle class and feature incremental learning in vertical federated learning, reporting large gains over baselines.