论文: M. Reimer, I. Cases, R. Ajemian, M. Liu, I. Rish, Y. Tu, G. Tesauro, Learning to learn without forgetting by maximizing transfer and minimizing interference, in: ICLR, 2019.
提出了Meta-Experience Replay (MER) 模型,将经验重放与基于优化的元学习相结合,基于Reptile元学习模型,以最大限度地转移和最小化基于未来梯度的干扰
【原创,转载需标明出处】论文解析(内含论文原文、代码链接):Notion – The all-in-one workspace for your notes, tasks, wikis, and databases.
代码注释(分别针对 Task-IL 任务增量和 Class-IL 类增量场景下):https://blog.csdn.net/qq_41933740/article/details/120484992