Personalized Federated Domain-Incremental Learning based on Adaptive Knowledge Matching

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Li, Yichen, Xu, Wenchao, Wang, Haozhao, Li, Ruixuan, Qi, Yining, Guo, Jingcai
Formato: Preprint
Publicado: 2024
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866914876199272448
author Li, Yichen
Xu, Wenchao
Wang, Haozhao
Li, Ruixuan
Qi, Yining
Guo, Jingcai
author_facet Li, Yichen
Xu, Wenchao
Wang, Haozhao
Li, Ruixuan
Qi, Yining
Guo, Jingcai
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2407_05005
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Personalized Federated Domain-Incremental Learning based on Adaptive Knowledge Matching
Li, Yichen
Xu, Wenchao
Wang, Haozhao
Li, Ruixuan
Qi, Yining
Guo, Jingcai
Machine Learning
Distributed, Parallel, and Cluster Computing
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.
title Personalized Federated Domain-Incremental Learning based on Adaptive Knowledge Matching
topic Machine Learning
Distributed, Parallel, and Cluster Computing
url https://arxiv.org/abs/2407.05005