用边界数据增强记忆，让模型学完新任务不丢旧知识。
Continual Learning with Support Boundary Experience Blending
- 在特征空间注入噪声生成边界附近数据，强化决策边界。
- 在多个数据集上提升准确率，最高达14%。
- 适合长期学习新任务且怕遗忘的场景。
持续学习旨在缓解模型在顺序任务训练中出现的灾难性遗忘问题。常用方法经验回放（ER）虽存储历史样本，但仅稀疏近似数据分布，导致决策边界脆弱且简化。本文提出支持边界数据（SBD），通过类差分隐私的噪声注入到隐层特征，生成靠近决策边界的表征，隐式正则化边界。基于此，提出经验融合（EB）框架，通过双模型聚合策略联合训练真实样本与SBD。EB包含：（1）隐空间噪声注入生成支持边界数据；（2）端到端训练同时利用真实样本与SBD。相比传统经验回放，SBD丰富了决策边界附近的特征空间，提升持续学习的稳定性与鲁棒性。在CIFAR-10、CIFAR-100、Tiny ImageNet和ImageNet1K上的大量实验表明，准确率分别提升10%、6%、14%、2%。
原文摘要 · Abstract (English)
Continual learning (CL) seeks to mitigate catastrophic forgetting when models are trained with sequential tasks. A common approach, experience replay (ER), stores past exemplars but only sparsely approximates the data distribution, yielding fragile and oversimplified decision boundaries. We address this limitation by introducing Support Boundary Data (SBD), generated via differential-privacy-inspired noise into latent features to create boundary-adjacent representations that implicitly regularize decision boundaries. Building on this idea, we propose Experience Blending (EB), a framework that jointly trains on exemplars and SBD through a dual-model aggregation strategy. EB has two components: (1) latent-space noise injection to generate support boundary data, and (2) end-to-end training that jointly leverages exemplars and SBD. Unlike standard experience replay, SBD enriches the feature space near decision boundaries, leading to more stable and robust continual learning. Extensive experiments on CIFAR-10, CIFAR-100, Tiny ImageNet, and ImageNet1K demonstrate consistent accuracy improvements of 10%, 6%, 14%, 2%, respectively.
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。