Bao Kexin, Lin Fanzhao, Wang Zichen, Li Yong, Zeng Dan, Ge Shiming
Institute of Information Engineering, Chinese Academy of Sciences, Beijing, 100092, China; School of Cyber Security, University of Chinese Academy of Sciences, Beijing, 100049, China.
NARI Technology Company Limited, Nanjing, 210000, Jiangsu, China.
Neural Netw. 2025 Jul 10;192:107724. doi: 10.1016/j.neunet.2025.107724.
Few-shot class-incremental learning (FSCIL) aims to continually adapt a model on a limited number of new-class examples, facing two well-known challenges: catastrophic forgetting and overfitting to new classes. Existing methods tend to freeze more parts of network components and finetune others with an extra memory during incremental sessions. These methods emphasize preserving prior knowledge to ensure proficiency in recognizing old classes, thereby mitigating catastrophic forgetting. Meanwhile, constraining fewer parameters can help in overcoming overfitting with the assistance of prior knowledge. Following previous methods, we retain more prior knowledge and propose a prior knowledge-infused neural network (PKI) to facilitate FSCIL. PKI consists of a backbone, an ensemble of projectors, a classifier, and an extra memory. In each incremental session, we build a new projector and add it to the ensemble. Subsequently, we finetune the new projector and the classifier jointly with other frozen network components, ensuring the rich prior knowledge is utilized effectively. By cascading projectors, PKI integrates prior knowledge accumulated from previous sessions and learns new knowledge flexibly, which helps to recognize old classes and efficiently learn new classes. Further, to reduce the resource consumption associated with keeping many projectors, we design two variants of the prior knowledge-infused neural network (PKIV-1 and PKIV-2) to trade off a balance between resource consumption and performance by reducing the number of projectors. Extensive experiments on three popular benchmarks demonstrate that our approach outperforms state-of-the-art methods.
少样本类别增量学习（FSCIL）旨在面对两个众所周知的挑战，即灾难性遗忘和对新类别的过拟合，在有限数量的新类示例上持续调整模型。现有方法在增量训练阶段往往会冻结网络组件的更多部分，并使用额外的内存对其他部分进行微调。这些方法强调保留先验知识以确保在识别旧类别方面的熟练度，从而减轻灾难性遗忘。同时，约束较少的参数有助于在先验知识的帮助下克服过拟合。遵循先前的方法，我们保留更多的先验知识，并提出一种注入先验知识的神经网络（PKI）来促进FSCIL。PKI由一个主干网络、一组投影器、一个分类器和一个额外的内存组成。在每个增量训练阶段，我们构建一个新的投影器并将其添加到投影器组中。随后，我们将新的投影器和分类器与其他冻结的网络组件一起进行联合微调，确保有效地利用丰富的先验知识。通过级联投影器，PKI整合了从先前训练阶段积累的先验知识，并灵活地学习新知识，这有助于识别旧类别并高效地学习新类别。此外，为了减少与保留多个投影器相关的资源消耗，我们设计了两种注入先验知识的神经网络变体（PKIV - 1和PKIV - 2），通过减少投影器的数量来在资源消耗和性能之间进行权衡。在三个流行基准上进行的广泛实验表明，我们的方法优于现有最先进的方法。
