Personalized Federated Domain-Incremental Learning based on Adaptive Knowledge Matching

TL;DR

This paper introduces pFedDIL, a personalized federated domain-incremental learning method that adaptively matches knowledge between tasks, improving accuracy by up to 14.35% over existing methods.

Contribution

The paper proposes an adaptive knowledge matching approach for personalized FDIL, enabling clients to select appropriate models based on task correlations and share parameters for better knowledge transfer.

Findings

pFedDIL outperforms state-of-the-art methods by up to 14.35% in accuracy.

Clients adaptively choose models based on task correlation, enhancing learning efficiency.

Sharing partial parameters reduces model complexity while maintaining performance.

Abstract

This paper focuses on Federated Domain-Incremental Learning (FDIL) where each client continues to learn incremental tasks where their domain shifts from each other. We propose a novel adaptive knowledge matching-based personalized FDIL approach (pFedDIL) which allows each client to alternatively utilize appropriate incremental task learning strategy on the correlation with the knowledge from previous tasks. More specifically, when a new task arrives, each client first calculates its local correlations with previous tasks. Then, the client can choose to adopt a new initial model or a previous model with similar knowledge to train the new task and simultaneously migrate knowledge from previous tasks based on these correlations. Furthermore, to identify the correlations between the new task and previous tasks for each client, we separately employ an auxiliary classifier to each target classification model and propose sharing partial parameters between the target classification model and the auxiliary classifier to condense model parameters. We conduct extensive experiments on several datasets of which results demonstrate that pFedDIL outperforms state-of-the-art methods by up to 14.35\% in terms of average accuracy of all tasks.