arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2003.03196cs.LGstat.ML
收藏

带加权迁移的联邦持续学习

Federated Continual Learning with Weighted Inter-client Transfer

Korea Advanced Institute of Science and Technology (KAIST)(韩国科学技术院);AITRICS

Jaehong Yoon, Wonyong Jeong, Giwoong Lee, Eunho Yang, Sung Ju Hwang

摘要:

人们对持续学习和联邦学习的兴趣激增,这两种学习在现实世界的深度神经网络中都很重要。然而,对于每个客户端从私人本地数据流中学习一系列任务的场景,人们几乎没有进行过研究。这种联合持续学习的问题对持续学习提出了新的挑战,例如利用来自其他客户端的知识,同时防止不相关知识的干扰。为了解决这些问题,我们提出了一种新的联邦持续学习框架,联邦加权客户端间传输(FedWeIT),它将网络权重分解为全局联邦参数和稀疏的任务特定参数,每个客户端从其他客户端接收选择性的知识,通过采取加权组合的任务特定参数。FedWeIT最大限度地减少了不兼容任务之间的干扰,并允许在学习过程中跨客户进行积极的知识转移。我们验证了我们的FedWeIT对现有的联邦学习和持续学习方法在不同程度的任务相似性的客户端,我们的模型显着优于他们的通信成本大幅降低。代码可在www.example.com上获得

英文摘要:

There has been a surge of interest in continual learning and federated learning, both of which are important in deep neural networks in real-world scenarios. Yet little research has been done regarding the scenario where each client learns on a sequence of tasks from a private local data stream. This problem of federated continual learning poses new challenges to continual learning, such as utilizing knowledge from other clients, while preventing interference from irrelevant knowledge. To resolve these issues, we propose a novel federated continual learning framework, Federated Weighted Inter-client Transfer (FedWeIT), which decomposes the network weights into global federated parameters and sparse task-specific parameters, and each client receives selective knowledge from other clients by taking a weighted combination of their task-specific parameters. FedWeIT minimizes interference between incompatible tasks, and also allows positive knowledge transfer across clients during learning. We validate our FedWeIT against existing federated learning and continual learning methods under varying degrees of task similarity across clients, and our model significantly outperforms them with a large reduction in the communication cost. Code is available at https://github.com/wyjeong/FedWeIT

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑