PIP: Prototypes-Injected Prompt for Federated Class Incremental Learning

TL;DR

The paper introduces PIP, a novel rehearsal-free federated class incremental learning method that uses prototypes-injected prompts, demonstrating significant performance improvements and robustness over existing methods across multiple datasets.

Contribution

PIP is the first to integrate prototype injection into prompt learning for FCIL, reducing communication costs and exemplars while improving accuracy and robustness.

Findings

Outperforms state-of-the-art methods by up to 33% on CIFAR100, MiniImageNet, and TinyImageNet.

Requires fewer local clients and global rounds, enhancing efficiency.

Demonstrates robustness across different task sizes.

Abstract

Federated Class Incremental Learning (FCIL) is a new direction in continual learning (CL) for addressing catastrophic forgetting and non-IID data distribution simultaneously. Existing FCIL methods call for high communication costs and exemplars from previous classes. We propose a novel rehearsal-free method for FCIL named prototypes-injected prompt (PIP) that involves 3 main ideas: a) prototype injection on prompt learning, b) prototype augmentation, and c) weighted Gaussian aggregation on the server side. Our experiment result shows that the proposed method outperforms the current state of the arts (SOTAs) with a significant improvement (up to 33%) in CIFAR100, MiniImageNet and TinyImageNet datasets. Our extensive analysis demonstrates the robustness of PIP in different task sizes, and the advantage of requiring smaller participating local clients, and smaller global rounds. For further study, source codes of PIP, baseline, and experimental logs are shared publicly in https://github.com/anwarmaxsum/PIP.