Federated Continual Learning

Follow this topic Ask about this topic

Federated continual learning studies how clients can learn from evolving tasks or data distributions without sharing raw data or losing previously acquired knowledge. Common approaches include replay buffers and synthetic replay, prototypes and distillation, parameter-importance regularization, and gradient or feature-subspace methods that limit interference; some systems also use client clustering, personalized models, or asynchronous updates. Methods are evaluated on class- and domain-incremental image benchmarks as well as medical, language, multimodal, and mobile settings, where client data and task order may differ. A central tension is balancing adaptation and retention under limited memory, communication, and privacy, while heterogeneous and task-free settings make results difficult to compare and expose weaknesses that average accuracy can hide.

Latest papers

37 papers7 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

MLFCIL: A Multi-Level Forgetting Mitigation Framework for Federated Class-Incremental Learning in LEO Satellites
Mar 14, 2026Heng Zhang, Xiaohong Deng, Sijing Duan +4Remote Sensing Scene Classification Federated Continual Learning
School of Information Engineering, Gannan University of Science and Technology, Ganzhou 341000, China · Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China · Department of Electrical and Computer Engineering, Queen University, Kingston, ON K7L 3N6, Canada +2
Catastrophic Forgetting Resilient One-Shot Incremental Federated Learning
Feb 19, 2026Obaidullah Zaland, Zulfiqar Ahmad Khan, Monowar BhuyanFederated Learning Federated Continual Learning
Department of Computing Science, Umeå University, Umeå, SE-90187, Sweden
Spatiotemporal Continual Federated Learning for Agentic Multi-UAV Edge Networks: Mitigating Catastrophic Forgetting
Jan 29, 2026Chuan-Chi LaiMulti-Objective Reinforcement Learning Federated RL
Department of Communications Engineering, National Chung Cheng University, Minxiong Township, Chiayi County 621301, Taiwan · Advanced Institute of Manufacturing with High-tech Innovations (AIM-HI), National Chung Cheng University, Minxiong Township, Chiayi County 621301, Taiwan
Knowledge-Aware Evolution for Task-Free Streaming Federated Continual Learning with Arbitrary Class Overlap
Jan 27, 2026Sixing Tan, Xianmin LiuContinual Learning Federated Continual Learning
Faculty of Computing Harbin Institute of Technology
Task-Agnostic Federated Continual Learning via Replay-Free Gradient Projection
Sep 25, 2025Seohyeon Cha, Huancheng Chen, Haris VikaloFederated Continual Learning Gradient Projection
Department of Electrical and Computer Engineering The University of Texas at Austin
Asynchronous Federated Continual Segmentation with Evolving Clients and Label Spaces
Mar 19, 2025Can Peng, Qianhui Men, Pramit Saha +5Federated Learning Aggregation Federated Continual Learning
University of Oxford, Oxford, United Kingdom · University of Bristol, Bristol, United Kingdom
Federated Class-Incremental Learning with Hierarchical Generative Prototypes
Jun 4, 2024Riccardo Salami, Pietro Buzzega, Matteo Mosconi +2Class-Incremental Learning Prototype-Based Classification
AImageLab - University of Modena and Reggio Emilia, Modena, Italy

Mar 14, 2026J/K move · Enter open · S save

MLFCIL: A Multi-Level Forgetting Mitigation Framework for Federated Class-Incremental Learning in LEO Satellites

Heng Zhang, Xiaohong Deng, Sijing Duan, Wu Ouyang+3

School of Information Engineering, Gannan University of Science and Technology, Ganzhou 341000, China · Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China · Department of Electrical and Computer Engineering, Queen University, Kingston, ON K7L 3N6, Canada+2

Remote Sensing Scene Classification Federated Continual Learning Federated Learning Catastrophic Forgetting

Open paper AskPDF arXiv

Low-Earth-orbit (LEO) satellite constellations are increasingly performing on-board computing. However, the continuous emergence of new classes under strict memory and communication constraints poses major challenges for collaborative training. Federated class-incremental learning (FCIL) enables distributed incremental learning without sharing raw data, but faces three LEO-specific challenges: non-independent and identically distributed data heterogeneity caused by orbital dynamics, amplified catastrophic forgetting during aggregation, and the need to balance stability and plasticity under limited resources. To tackle these challenges, we propose MLFCIL, a multi-level forgetting mitigation framework that decomposes catastrophic forgetting into three sources and addresses them at different levels: class-reweighted loss to reduce local bias, knowledge distillation with feature replay and prototype-guided drift compensation to preserve cross-task knowledge, and class-aware aggregation to mitigate forgetting during federation. In addition, we design a dual-granularity coordination strategy that combines round-level adaptive loss balancing with step-level gradient projection to further enhance the stability-plasticity trade-off. Experiments on the NWPU-RESISC45 dataset show that MLFCIL significantly outperforms baselines in both accuracy and forgetting mitigation, while introducing minimal resource overhead.