基于原型示例压缩的内存高效持续学习
Memory-efficient Continual Learning with Prototypical Exemplar Condensation
AI总结：
本文提出通过合成和存储原型示例压缩内存，结合扰动增强机制提升持续学习性能，在大规模数据集和多任务场景中表现更优。
摘要：
基于回忆的持续学习（CL）通过维护先前任务的子集样本进行重放以缓解灾难性遗忘。现有研究主要聚焦于通过coreset选择策略优化内存存储。尽管这些方法有效，但通常需要存储每个类别的大量样本（SPC），往往超过20个以维持满意性能。在本文中，我们提出进一步压缩内存足迹，通过合成和存储原型示例，这些示例在通过特征提取器后可形成代表性原型。由于其代表性，这些示例使模型能够在少量样本上保留先前知识的同时保持隐私。此外，我们引入基于扰动的增强机制，在训练期间生成先前数据的合成变体，从而提升CL性能。在广泛使用的基准数据集和设置上的大量评估表明，所提出的算法在现有基线相比，在大规模数据集和高任务数场景中表现更优。
英文摘要：
Rehearsal-based continual learning (CL) mitigates catastrophic forgetting by maintaining a subset of samples from previous tasks for replay. Existing studies primarily focus on optimizing memory storage through coreset selection strategies. While these methods are effective, they typically require storing a substantial number of samples per class (SPC), often exceeding 20, to maintain satisfactory performance. In this work, we propose to further compress the memory footprint by synthesizing and storing prototypical exemplars, which can form representative prototypes when passed through the feature extractor. Owing to their representative nature, these exemplars enable the model to retain previous knowledge using only a small number of samples while preserving privacy. Moreover, we introduce a perturbation-based augmentation mechanism that generates synthetic variants of previous data during training, thereby enhancing CL performance. Extensive evaluations on widely used benchmark datasets and settings demonstrate that the proposed algorithm achieves superior performance compared to existing baselines, particularly in scenarios involving large-scale datasets and a high number of tasks.
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。
