DOLFIN提升联邦持续学习稳定性与效率，防止遗忘且通信开销低。
DOLFIN: Balancing Stability and Plasticity in Federated Continual Learning
- 用低秩适配器+双梯度记忆，实现轻量高效更新
- 在多个数据集上超越6种基线方法，准确率更高
- 适合隐私敏感的分布式持续学习场景
联邦持续学习（FCL）使模型能在多个分布式客户端上持续学习新任务，保护隐私并避免遗忘旧知识。然而，现有方法难以兼顾性能、隐私保护和通信效率。本文提出一种基于视觉变压器的分布式在线低秩适配器方法DOLFIN，通过低秩适配器实现极小通信开销，并引入双梯度投影记忆（DualGPM）防止遗忘。在CIFAR-100、ImageNet-R、ImageNet-A和CUB-200四个数据集上，于两种狄利克雷异构设置下评估，DOLFIN始终优于六种强基线方法，在最终平均准确率上领先，同时保持相同内存占用。正交低秩适配器为隐私保护的联邦持续学习提供了有效且可扩展的解决方案。
原文摘要 · Abstract (English)
Federated continual learning (FCL) enables models to learn new tasks across multiple distributed clients, protecting privacy and without forgetting previously acquired knowledge. However, current methods face challenges balancing performance, privacy preservation, and communication efficiency. We introduce a Distributed Online LoRA for Federated INcremental learning method DOLFIN, a novel approach combining Vision Transformers with low-rank adapters designed to efficiently and stably learn new tasks in federated environments. Our method leverages LoRA for minimal communication overhead and incorporates DualGradient Projection Memory (DualGPM) to prevent forgetting. Evaluated on CIFAR-100, ImageNet-R, ImageNet-A, and CUB-200 under two Dirichlet heterogeneity settings, DOLFIN consistently surpasses six strong baselines in final average accuracy while matching their memory footprint. Orthogonal low-rank adapters offer an effective and scalable solution for privacy-preserving continual learning in federated settings.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。