解决少样本增量学习中旧知识与新知识的冲突问题。
Consistency-Driven Calibration and Matching for Few-Shot Class-Incremental Learning
- 通过记忆感知的原型校准，提升特征概念中心一致性。
- 动态结构匹配实现跨会话几何最优对齐，最高提升3.41%准确率。
- 无需类别数量先验，适合开放世界场景下的持续学习应用。
少样本类增量学习（FSCIL）对于适应复杂开放世界环境至关重要。现有基于展望的学习空间构建方法难以平衡新旧知识，因原型偏差和刚性结构限制了嵌入空间的表达能力。本文从特征-结构双重一致性角度重新思考优化难题，提出一致性驱动的校准与匹配（ConCM）框架，系统缓解FSCIL中的知识冲突。受海马体联想记忆启发，设计记忆感知的原型校准机制，从基础类提取通用语义属性并重用于新类，增强特征概念中心的一致性；进一步提出动态结构匹配，自适应对齐校准特征至会话特定最优流形空间，确保跨会话结构一致性。该过程无需类别数量先验，理论上保证几何最优与最大匹配性。在mini-ImageNet、CIFAR100和CUB200等大规模基准上，ConCM达到领先性能，增量会话中谐波准确率最高提升3.41%。代码已公开于：https://github.com/wire-wqz/ConCM。
原文摘要 · Abstract (English)
Few-Shot Class Incremental Learning (FSCIL) is crucial for adapting to the complex open-world environments. Contemporary prospective learning-based space construction methods struggle to balance old and new knowledge, as prototype bias and rigid structures limit the expressive capacity of the embedding space. Different from these strategies, we rethink the optimization dilemma from the perspective of feature-structure dual consistency, and propose a Consistency-driven Calibration and Matching (ConCM) framework that systematically mitigates the knowledge conflict inherent in FSCIL. Specifically, inspired by hippocampal associative memory, we design a memory-aware prototype calibration that extracts generalized semantic attributes from base classes and reintegrates them into novel classes to enhance the conceptual center consistency of features. Further, to consolidate memory associations, we propose dynamic structure matching, which adaptively aligns the calibrated features to a session-specific optimal manifold space, ensuring cross-session structure consistency. This process requires no class-number priors and is theoretically guaranteed to achieve geometric optimality and maximum matching. On large-scale FSCIL benchmarks including mini-ImageNet, CIFAR100 and CUB200, ConCM achieves state-of-the-art performance, with harmonic accuracy gains of up to 3.41% in incremental sessions. Code is available at: https://github.com/wire-wqz/ConCM
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。