不微调参数，用类比生成新类别权重，提升小样本增量学习性能
Learn by Reasoning: Analogical Weight Generation for Few-Shot Class-Incremental Learning
- 基于人类类比思维设计权重生成机制，无需微调
- 在miniImageNet等数据集上平均准确率超当前最优方法
- 适合需要持续学习新类别的实际应用
少样本增量学习（FSCIL）使模型能在少量新数据下学习新类别，同时保持对旧类别的性能。传统方法需在有限新数据上微调参数，且新旧知识利用存在割裂。受人类大脑类比学习机制启发，我们提出一种新型类比生成方法。该方法包含脑启发类比生成器（BiAG），可在增量阶段无需参数微调，直接从已有类别生成新类别权重。BiAG由三部分构成：权重自注意力模块（WSA）、权重与原型类比注意力模块（WPAA）和语义转换模块（SCM）。SCM利用神经坍缩理论实现语义转换，WSA补充新类别权重，WPAA计算类比关系以生成新权重。在miniImageNet、CUB-200和CIFAR-100上的实验表明，该方法在最终准确率和平均准确率上均优于现有最先进方法。
原文摘要 · Abstract (English)
Few-shot class-incremental Learning (FSCIL) enables models to learn new classes from limited data while retaining performance on previously learned classes. Traditional FSCIL methods often require fine-tuning parameters with limited new class data and suffer from a separation between learning new classes and utilizing old knowledge. Inspired by the analogical learning mechanisms of the human brain, we propose a novel analogical generative method. Our approach includes the Brain-Inspired Analogical Generator (BiAG), which derives new class weights from existing classes without parameter fine-tuning during incremental stages. BiAG consists of three components: Weight Self-Attention Module (WSA), Weight & Prototype Analogical Attention Module (WPAA), and Semantic Conversion Module (SCM). SCM uses Neural Collapse theory for semantic conversion, WSA supplements new class weights, and WPAA computes analogies to generate new class weights. Experiments on miniImageNet, CUB-200, and CIFAR-100 datasets demonstrate that our method achieves higher final and average accuracy compared to SOTA methods.
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。