提出参数相对重要性度量，实现无任务标识下的持续学习
Relative Parameter Importance in Task-Agnostic Replay-Free Continual Learning
- 基于参数对当前与历史任务的相对重要性动态调节更新
- 在文本分类任务上超越现有最优方法，提升稳定性和知识迁移
- 适合研究持续学习中权重保护与知识传递机制的读者
实现深度神经网络的持续学习需平衡稳定性与可塑性并支持知识迁移。本文研究在以下约束下的离线学习算法：(I) 无法访问先前任务的训练数据；(II) 推理时无任务标识。提出一种新度量——相对参数重要性，用于衡量每个参数相对于当前和过往任务的重要性。高相对重要性的参数被视为维持旧任务稳定的关键，因而被强正则化；低相对重要性的参数则允许更自由更新。与现有方法不同，本方法可在旧任务重要性高的参数具有低相对重要性时仍允许其更新，从而实现向后知识迁移，同时缓解稳定性-可塑性权衡问题。在类别增量与领域增量文本分类任务上均取得优于现有最优方法的结果，并为扩展至文本生成提供启示。代码已公开于：https://github.com/itsmemala/LACL
原文摘要 · Abstract (English)
Achieving continual learning (CL) with deep neural networks requires balancing stability and plasticity while enabling knowledge transfer. In this work, we focus on offline learning algorithms under the constraints: (I) no access to training data from prior tasks (II) no access to task-id at inference time. We introduce a novel measure, the relative parameter-importance, which measures the relative importance of each parameter with respect to both the current and past tasks. Parameters with high relative importance are interpreted as more important for maintaining past-task stability and thus heavily regularised, whereas parameters with low relative-importance are allowed to be more freely updated. Unlike existing methods, our approach allows the update of parameters with high past-task importance when they have low relative-importance, thus enabling backward knowledge transfer in addition to tackling the stability-plasticity trade-off. We demonstrate improvements against state-of-the-art CL methods on both class-incremental and domain-incremental learning text classification problems and provide insights for extending our method to text generation problems. Code available at: https://github.com/itsmemala/LACL
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。