用低秩适配实现实时无任务持续学习，避免遗忘且无需记忆缓冲。
Online-LoRA: Task-free Online Continual Learning via Low Rank Adaptation
- 通过在线权重正则化识别关键参数并加固
- 自动检测数据分布变化，实现无任务边界适应
- 适用于多种ViT模型，性能超越当前最优方法
灾难性遗忘是在线持续学习（OCL）中的重大挑战，尤其在无明确任务边界的非平稳数据流中更为突出。传统基于重放缓冲区的方法受限于内存与隐私问题。为此，本文提出Online-LoRA框架，支持对预训练Vision Transformer（ViT）模型进行实时微调，克服重放缓冲区的局限，并发挥预训练模型的优势。核心贡献在于一种新颖的在线权重正则化策略，可识别并巩固重要模型参数；同时利用损失值的训练动态实现数据分布偏移的自动识别。在多个无任务持续学习场景及基准数据集（包括CIFAR-100、ImageNet-R、ImageNet-S、CUB-200和CORe50）上的大量实验表明，Online-LoRA能稳健适配多种ViT架构，且性能优于现有最先进方法。代码将公开于：https://github.com/Christina200/Online-LoRA-official.git。
原文摘要 · Abstract (English)
Catastrophic forgetting is a significant challenge in online continual learning (OCL), especially for non-stationary data streams that do not have well-defined task boundaries. This challenge is exacerbated by the memory constraints and privacy concerns inherent in rehearsal buffers. To tackle catastrophic forgetting, in this paper, we introduce Online-LoRA, a novel framework for task-free OCL. Online-LoRA allows to finetune pre-trained Vision Transformer (ViT) models in real-time to address the limitations of rehearsal buffers and leverage pre-trained models' performance benefits. As the main contribution, our approach features a novel online weight regularization strategy to identify and consolidate important model parameters. Moreover, Online-LoRA leverages the training dynamics of loss values to enable the automatic recognition of the data distribution shifts. Extensive experiments across many task-free OCL scenarios and benchmark datasets (including CIFAR-100, ImageNet-R, ImageNet-S, CUB-200 and CORe50) demonstrate that Online-LoRA can be robustly adapted to various ViT architectures, while achieving better performance compared to SOTA methods. Our code will be publicly available at: https://github.com/Christina200/Online-LoRA-official.git.
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。