FedRAN achieves up to 4.8 pp higher accuracy in federated continual learning while using 30-122× less per-client communication by transmitting truncated-SVD summaries of random-feature Gram matrices and performing closed-form ridge classification after two-level QR-SVD merging.
Fed- cprompt: Contrastive prompt for rehearsal-free federated continual learning
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
cs.LG 2verdicts
UNVERDICTED 2representative citing papers
Fed-TaLoRA uses task-agnostic low-rank residual adaptation with post-aggregation calibration to enable efficient federated continual fine-tuning across sequential tasks under non-IID conditions.
citing papers explorer
-
Accurate and Resource-Efficient Federated Continual Learning
FedRAN achieves up to 4.8 pp higher accuracy in federated continual learning while using 30-122× less per-client communication by transmitting truncated-SVD summaries of random-feature Gram matrices and performing closed-form ridge classification after two-level QR-SVD merging.
-
Task-agnostic Low-rank Residual Adaptation for Efficient Federated Continual Fine-Tuning
Fed-TaLoRA uses task-agnostic low-rank residual adaptation with post-aggregation calibration to enable efficient federated continual fine-tuning across sequential tasks under non-IID conditions.