arXivDaily arXiv每日学术速递 周一至周五更新
​通知:联邦节假日调休:哥伦布日 / 原住民日(Columbus Day / Indigenous Peoples' Day),今日(13日)无更新。
arXiv 2307.04869cs.LGcs.AIcs.CV

Fed-CPrompt:面向无重放联邦持续学习的对比提示

Fed-CPrompt: Contrastive Prompt for Rehearsal-Free Federated Continual Learning

Michigan Technological University(密歇根理工大学);University of Houston(休斯顿大学)

Gaurav Bagwe, Xiaoyong Yuan, Miao Pan, Lan Zhang

AI总结:

针对无重放联邦持续学习中的灾难性遗忘,提出基于提示学习的Fed-CPrompt,通过异步提示学习和对比持续损失处理异步任务与异构数据,实现通信高效的最优性能。

摘要:

联邦持续学习(FCL)随时间从分布在各个客户端的机密数据集中学习增量任务。本文聚焦于无重放FCL,由于无法访问历史任务数据,在学习新任务时存在严重的遗忘问题。为解决此问题,我们基于提示学习技术提出Fed-CPrompt,以通信高效的方式获取任务特定提示。Fed-CPrompt引入两个关键组件:异步提示学习和对比持续损失,分别处理FCL中的异步任务到达和异构数据分布。大量实验表明Fed-CPrompt在实现最先进的无重放FCL性能方面的有效性。

英文摘要:

Federated continual learning (FCL) learns incremental tasks over time from confidential datasets distributed across clients. This paper focuses on rehearsal-free FCL, which has severe forgetting issues when learning new tasks due to the lack of access to historical task data. To address this issue, we propose Fed-CPrompt based on prompt learning techniques to obtain task-specific prompts in a communication-efficient way. Fed-CPrompt introduces two key components, asynchronous prompt learning, and contrastive continual loss, to handle asynchronous task arrival and heterogeneous data distributions in FCL, respectively. Extensive experiments demonstrate the effectiveness of Fed-CPrompt in achieving SOTA rehearsal-free FCL performance.

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。

↑