基于网络拆分与梦境元加权模型融合的持续学习
Continual Learning through Networks Splitting and Merging with Dreaming-Meta-Weighted Model Fusion
AI总结：
针对持续学习中稳定性与可塑性难平衡的问题，提出Split2MetaFusion两阶段方法，通过拆分快慢模型、结合TPNSP优化器和梦境元加权融合策略，在无需旧数据集的情况下实现更优的性能权衡。
摘要：
在持续学习场景中，平衡网络的稳定性与可塑性是一项挑战，因为模型更新会损害稳定性，却能提升可塑性。现有研究通常更侧重稳定性，限制后续任务的学习可塑性，以避免已学知识发生灾难性遗忘。与之不同，我们提出一种名为Split2MetaFusion的持续学习方法，通过“拆分-元加权融合”两阶段策略实现更优的权衡。在拆分阶段，依次学习具备更优稳定性的慢模型和具备更优可塑性的快模型；随后通过自适应融合两个模型，同时保留稳定性与可塑性。为此，我们设计了名为Task-Preferred Null Space Projector（任务偏好零空间投影器，简称TPNSP）的优化器，用于慢学习过程以缩小融合差距。为实现更优的模型融合，我们进一步设计了Dreaming-Meta-Weighted（梦境元加权）融合策略，无需使用过往数据集，就能更好地同时维持新旧知识。实验结果与分析表明，该方法在维持网络稳定性和保持可塑性方面具有优越性，我们的代码将公开发布。
英文摘要：
It's challenging to balance the networks stability and plasticity in continual learning scenarios, considering stability suffers from the update of model and plasticity benefits from it. Existing works usually focus more on the stability and restrict the learning plasticity of later tasks to avoid catastrophic forgetting of learned knowledge. Differently, we propose a continual learning method named Split2MetaFusion which can achieve better trade-off by employing a two-stage strategy: splitting and meta-weighted fusion. In this strategy, a slow model with better stability, and a fast model with better plasticity are learned sequentially at the splitting stage. Then stability and plasticity are both kept by fusing the two models in an adaptive manner. Towards this end, we design an optimizer named Task-Preferred Null Space Projector(TPNSP) to the slow learning process for narrowing the fusion gap. To achieve better model fusion, we further design a Dreaming-Meta-Weighted fusion policy for better maintaining the old and new knowledge simultaneously, which doesn't require to use the previous datasets. Experimental results and analysis reported in this work demonstrate the superiority of the proposed method for maintaining networks stability and keeping its plasticity. Our code will be released.
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。
