在线曲率感知回放：利用二阶信息进行在线持续学习
Online Curvature-Aware Replay: Leveraging $\mathbf{2^{nd}}$ Order Information for Online Continual Learning
AI总结：
本文提出在线曲率感知回放（OCAR）方法，将基于回放的在线持续学习形式化为带KL约束的二阶在线联合优化，利用K-FAC近似的Fisher信息矩阵预处理梯度以稳定优化并防止遗忘，在多个基准上取得最优性能。
摘要：
在线持续学习（OCL）模型持续适应非平稳数据流，通常没有任务信息。这些设置很复杂，许多传统的持续学习方法会失败，而在线方法（主要基于回放）在任务转换后会遭受不稳定问题。为了解决这个问题，我们将基于回放的OCL形式化为一个二阶在线联合优化问题，并对回放数据施加显式的KL散度约束。我们提出在线曲率感知回放（OCAR）来解决该问题：一种利用损失的二阶信息的方法，使用Fisher信息矩阵（FIM）的K-FAC近似来对梯度进行预处理。FIM作为稳定器防止遗忘，同时加速非干扰方向上的优化。我们展示了如何将FIM的估计适应到持续学习设置中，为非独立同分布数据稳定二阶优化，揭示了Tikhonov正则化在稳定性-可塑性权衡中的作用。实证结果表明，OCAR在持续学习指标上优于最先进的方法，在三个不同的基准测试中在整个训练过程中实现了更高的平均准确率。
英文摘要：
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.
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。
