Abstract:As Web technology continues to develop, it has become increasingly common to use data stored on different clients. At the same time, federated learning has received widespread attention due to its ability to protect data privacy when let models learn from data which is distributed across various clients. However, most existing works assume that the client's data are fixed. In real-world scenarios, such an assumption is most likely not true as data may be continuously generated and new classes may also appear. To this end, we focus on the practical and challenging federated class-incremental learning (FCIL) problem. For FCIL, the local and global models may suffer from catastrophic forgetting on old classes caused by the arrival of new classes and the data distributions of clients are non-independent and identically distributed (non-iid). In this paper, we propose a novel method called Federated Class-Incremental Learning with PrompTing (FCILPT). Given the privacy and limited memory, FCILPT does not use a rehearsal-based buffer to keep exemplars of old data. We choose to use prompts to ease the catastrophic forgetting of the old classes. Specifically, we encode the task-relevant and task-irrelevant knowledge into prompts, preserving the old and new knowledge of the local clients and solving the problem of catastrophic forgetting. We first sort the task information in the prompt pool in the local clients to align the task information on different clients before global aggregation. It ensures that the same task's knowledge are fully integrated, solving the problem of non-iid caused by the lack of classes among different clients in the same incremental task. Experiments on CIFAR-100, Mini-ImageNet, and Tiny-ImageNet demonstrate that FCILPT achieves significant accuracy improvements over the state-of-the-art methods.

Federated Class-Incremental Learning with New-Class Augmented Self-Distillation

TARGET: Federated Class-Continual Learning Via Exemplar-Free Distillation

Privacy-Preserving Federated Class-Incremental Learning

Federated Class-Incremental Learning with Dynamic Feature Extractor Fusion

Data-Free Federated Class Incremental Learning with Diffusion-Based Generative Memory

Overcoming Spatial-Temporal Catastrophic Forgetting for Federated Class-Incremental Learning

FL-Clip: Bridging Plasticity and Stability in Pre-Trained Federated Class-Incremental Learning Models

FedINC: an Exemplar-Free Continual Federated Learning Framework with Small Labeled Data

Federated Class-Incremental Learning with Prompting

FedSKF: Selective Knowledge Fusion via Optimal Transport in Federated Class Incremental Learning

Federated Class-Incremental Learning with Hierarchical Generative Prototypes

FedProK: Trustworthy Federated Class-Incremental Learning via Prototypical Feature Knowledge Transfer

No One Left Behind: Real-World Federated Class-Incremental Learning

Overcoming Catastrophic Forgetting in Federated Class-Incremental Learning via Federated Global Twin Generator

Few-Shot Class-Incremental Learning Via Class-Aware Bilateral Distillation

Knowledge Lock: Overcoming Catastrophic Forgetting in Federated Learning

Learning by Imitating the Classics: Mitigating Class Imbalance in Federated Learning Via Simulated Centralized Learning

FedDistill: Global Model Distillation for Local Model De-Biasing in Non-IID Federated Learning

FedACL: Federated Multi-Distillation with Auxiliary Classification Layers in IoT.

Continual Federated Learning Based on Knowledge Distillation

FedTC: A Personalized Federated Learning Method with Two Classifiers