Class-incremental learning (CIL) has emerged as a means to learn new classes incrementally without catastrophic forgetting of previous classes. Recently, CIL has undergone a paradigm shift towards dynamic architectures due to their superior performance. However, these models are still limited by the following aspects: (i) Data augmentation (DA), which are tightly coupled with CIL, remains under-explored in dynamic architecture scenarios. (ii) Feature representation. The discriminativeness of dynamic feature are sub-optimal and possess potential for refinement. (iii) Classifier. The misalignment between dynamic feature and classifier constrains the capabilities of the model. To tackle the aforementioned drawbacks, we propose the Dynamic Feature Learning and Matching (DFLM) model in this paper from above three perspectives. Specifically, we firstly introduce class weight information and non-stationary functions to extend the mix DA method for dynamically adjusting the focus on memory during training. Then, von Mises-Fisher (vMF) classifier is employed to effectively model the dynamic feature distribution and implicitly learn their discriminative properties. Finally, the matching loss is proposed to facilitate the alignment between the learned dynamic features and the classifier by minimizing the distribution distance. Extensive experiments on CIL benchmarks validate that our proposed model achieves significant performance improvements over existing methods.

翻译：类增量学习（CIL）作为一种在避免灾难性遗忘旧类别的同时逐步学习新类别的方法应运而生。近年来，CIL 的研究范式已转向动态架构，因其性能更优。然而，这些模型仍受限于以下几个方面：(i) 数据增强（DA）与 CIL 紧密耦合，但在动态架构场景下尚未得到充分探索。(ii) 特征表示。动态特征的判别性仍非最优，存在进一步优化的空间。(iii) 分类器。动态特征与分类器之间的错位限制了模型的能力。针对上述不足，本文从上述三个角度提出了动态特征学习与匹配（DFLM）模型。具体而言，我们首先引入类别权重信息和非平稳函数，以扩展混合数据增强方法，从而在训练过程中动态调整对记忆的关注。然后，采用 von Mises-Fisher（vMF）分类器有效建模动态特征分布，并隐式学习其判别属性。最后，提出匹配损失，通过最小化分布距离来促进所学动态特征与分类器之间的对齐。在 CIL 基准上的大量实验表明，我们提出的模型相较于现有方法取得了显著的性能提升。