mlsanityData to 14 Oct 2026

Accurate and Resource-Efficient Federated Continual Learning

arXiv 2606.11480 ↗First posted 9 Jun 2026cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Federated continual learning (FCL) must learn from distributed task streams under limited resources, such as communication, computation, memory, and label availability. Existing FCL methods often rely on repeated local optimization, replay, and full supervision. Analytic alternatives avoid iterative training and replay, but using high-dimensional random features to improve accuracy requires a second-order feature statistic, the Gram matrix, which has a quadratic communication cost in the random feature size M. We propose FedRAN, a resource-aware analytic FCL framework that replaces gradient-based updates with compact random feature statistics. Each client transmits a truncated-SVD summary of its Gram matrix, reducing the dominant second-order upload from quadratic to linear in M for fixed rank. The server performs a two-level QR-SVD subspace merge, spatially across clients and temporally across tasks, and solves a ridge classifier in closed form. FedRAN further supports label scarcity through prototype-based pseudo-labeling. Across CIFAR-100, ImageNet-R, and VTAB datasets, FedRAN improves average accuracy by up to 4.8 percentage points over the strongest baseline, uses 30.6-121.8× less per-client communication than optimization-based FCL, and is 190.3× faster on average than gradient-based baselines; with only 20% labels, pseudo-labeling improves average accuracy by up to 6.61 points. These results show that FedRAN enables accurate and resource-efficient FCL under communication, computation, and label constraints. The source code is available at https://github.com/JebacyrilArockiaraj/Fed-RAN-SSL.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 37 of its 58 references that are in the map. First posted on arXiv 9 Jun 2026.

Builds on

37 of 58 references.

References · 37

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2020An Image is Worth 16x16 Words: Transformers for Image Recognition at Scalecited by 30,186
  3. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  4. 2016Overcoming catastrophic forgetting in neural networkscited by 4,787
  5. 2019Advances and Open Problems in Federated Learningcited by 3,046
  6. 2016Learning without Forgettingcited by 2,496
  7. 2016iCaRL: Incremental Classifier and Representation Learningcited by 2,333
  8. 2022DualPrompt: Complementary Prompting for Rehearsal-free Continual Learningcited by 642
  9. 2022CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learningcited by 450
  10. 2019A Large-scale Study of Representation Learning with the Visual Task Adaptation Benchmarkcited by 425
  11. 2021FedProto: Federated Prototype Learning across Heterogeneous Clientscited by 352
  12. 2023RanPAC: Random Projections and Pre-trained Models for Continual Learningcited by 164
25 more
  1. 2020Federated Continual Learning with Weighted Inter-client Transfercited by 161
  2. 2020Federated Semi-Supervised Learning with Inter-Client Consistency & Disjoint Learningcited by 112
  3. 2023TARGET: Federated Class-Continual Learning via Exemplar-Free Distillationcited by 65
  4. 2022ACIL: Analytic Class-Incremental Learning with Absolute Memorization and Privacy Protectioncited by 57
  5. 2023Better Generative Replay for Continual Federated Learningcited by 55
  6. 2015Frequent Directions : Simple and Deterministic Matrix Sketchingcited by 39
  7. 2024GACL: Exemplar-Free Generalized Analytic Continual Learningcited by 22
  8. 2023Federated Orthogonal Training: Mitigating Global Catastrophic Forgetting in Continual Federated Learningcited by 21
  9. 2024Random Representations Outperform Online Continually Learned Representationscited by 11
  10. 2024Accelerating Heterogeneous Federated Learning with Closed-form Classifierscited by 10
  11. 2024Handling Spatial-Temporal Data Heterogeneity for Federated Continual Learning via Tail Anchorcited by 9
  12. 2024LoRanPAC: Low-rank Random Features and Pre-trained Models for Bridging Theory and Practice in Continual Learningcited by 8
  13. 2025SacFL: Self-Adaptive Federated Continual Learning for Resource-Constrained End Devicescited by 8
  14. 2025AnaCP: Toward Upper-Bound Continual Learning via Analytic Contrastive Projectioncited by 7
  15. 2025Resource-Constrained Federated Continual Learning: What Does Matter?cited by 5
  16. 2026FLoRG: Federated Fine-tuning with Low-rank Gram Matrices and Procrustes Alignmentcited by 5
  17. 2024Covariances for Free: Exploiting Mean Distributions for Training-free Federated Learningcited by 4
  18. 2026Exact Federated Continual Unlearning for Ridge Heads on Frozen Foundation Modelscited by 4
  19. 2026The Gaussian-Head OFL Family: One-Shot Federated Learning from Client Global Statisticscited by 3
  20. 2025APFL: Analytic Personalized Federated Learning via Dual-Stream Least Squarescited by 2
  21. 2025C^2Prompt: Class-aware Client Knowledge Interaction for Federated Continual Learningcited by 2
  22. 2025Enhancing Federated Class-Incremental Learning via Spatial-Temporal Statistics Aggregationcited by 1
  23. 2025FedGTEA: Federated Class-Incremental Learning with Gaussian Task Embedding and Alignmentcited by 1
  24. 2025CATCHFed: Efficient Unlabeled Data Utilization for Semi-Supervised Federated Learning in Limited Labels Environmentscited by 1
  25. 2026Concatenated Matrix SVD: Compression Bounds, Incremental Approximation, and Error-Constrained Clusteringcited by 1

Built on by

Not cited yet.