用二阶信息增强在线持续学习的重放机制，提升稳定性与性能。
Online Curvature-Aware Replay: Leveraging $\mathbf{2^{nd}}$ Order Information for Online Continual Learning
- 基于FIM近似构建二阶优化框架，用K-FAC预条件梯度。
- 在三个基准上平均准确率超越当前最优方法，全程更稳定。
- 适合需要长期适应非平稳数据流的在线学习场景。
在线持续学习（OCL）模型需持续适应非平稳数据流，通常缺乏任务信息。传统方法在此类复杂场景中表现不佳，而主流基于重放的在线方法在任务切换后常出现不稳定性。本文将基于重放的OCL形式化为带显式KL散度约束的二阶在线联合优化问题。提出在线曲率感知重放（OCAR）：利用损失函数的二阶信息，通过K-FAC近似费雪信息矩阵（FIM）对梯度进行预条件处理。FIM作为稳定器可防止遗忘，并加速非干扰方向的优化。我们展示了如何在持续学习设定下适配FIM估计，稳定非独立同分布数据上的二阶优化，揭示了Tikhonov正则化在稳定-弹性权衡中的作用。实验表明，OCAR在三个不同基准上均优于现有最先进方法，在整个训练过程中保持更高平均准确率。
原文摘要 · Abstract (English)
Online Continual Learning (OCL) models continuously adapt to nonstationary data streams, usually without task information. These settings are complex and many traditional CL methods fail, while online methods (mainly replay-based) suffer from instabilities after the task shift. To address this issue, we formalize replay-based OCL as a second-order online joint optimization with explicit KL-divergence constraints on replay data. We propose Online Curvature-Aware Replay (OCAR) to solve the problem: a method that leverages second-order information of the loss using a K-FAC approximation of the Fisher Information Matrix (FIM) to precondition the gradient. The FIM acts as a stabilizer to prevent forgetting while also accelerating the optimization in non-interfering directions. We show how to adapt the estimation of the FIM to a continual setting stabilizing second-order optimization for non-iid data, uncovering the role of the Tikhonov regularization in the stability-plasticity tradeoff. Empirical results show that OCAR outperforms state-of-the-art methods in continual metrics achieving higher average accuracy throughout the training process in three different benchmarks.
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。