持续学习作为兼容位移下的共享流形延续
Continual Learning as Shared-Manifold Continuation Under Compatible Shift
AI总结：
本文提出基于共享流形延续的持续学习方法，通过稀疏重放、输出蒸馏等技术提升旧任务保留和表示保持，同时在新任务准确率上保持竞争力。
摘要：
持续学习方法通常通过参数正则化、输出匹配或重放先前示例来保留旧行为。这些策略可以减少遗忘，但并未直接规定潜在表示应如何演变。我们研究了一种更窄的几何替代方案，即在旧数据和新数据应保持在相同潜在支持的 regime 中，将持续学习视为共享流形的延续。我们在此观点下提出了支持保持流形 assimilation（SPMA）并评估了其几何保持变体 SPMA-OG，该方法结合了稀疏重放、输出蒸馏、关系几何保持、局部平滑和锚点分配正则化。在代表性的兼容位移 CIFAR10 和 Tiny-ImageNet 运行中，SPMA-OG 在旧任务保留和表示保持指标上优于稀疏重放基线，同时在新任务准确率上保持竞争力。在受控的合成 atlas-流形基准测试中，它实现了几乎完美的锚点几何保持，同时在新任务准确率上优于重放。这些结果提供了证据，表明在持续学习应保持共享潜在支持而非创建新支持时，几何感知的锚点正则化是一种有用的归纳偏置。
英文摘要：
Continual learning methods usually preserve old behavior by regularizing parameters, matching old outputs, or replaying previous examples. These strategies can reduce forgetting, but they do not directly specify how the latent representation should evolve. We study a narrower geometric alternative for the regime where old and new data should remain on the same latent support: continual learning as continuation of a shared manifold. We instantiate this view within Support-Preserving Manifold Assimilation (SPMA) and evaluate a geometry-preserving variant, SPMA-OG, that combines sparse replay, output distillation, relational geometry preservation, local smoothing, and chart-assignment regularization on old anchors. On representative compatible-shift CIFAR10 and Tiny-ImageNet runs, SPMA-OG improves over sparse replay baselines in old-task retention and representation-preservation metrics while remaining competitive on new-task accuracy. On a controlled synthetic atlas-manifold benchmark, it achieves near-perfect anchor-geometry preservation while also improving new-task accuracy over replay. These results provide evidence that geometry-aware anchor regularization is a useful inductive bias when continual learning should preserve a shared latent support rather than create a new one.
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。
