提出DCNet方法，解决无示例增量学习中类间混淆问题。
On the Discrimination and Consistency for Exemplar-Free Class Incremental Learning
- 将类别特征映射到超球面，实现类间正交分离。
- 通过补偿性训练动态调整监督强度，增强类内聚合。
- 适合需要持续学习新类且不存旧样本的场景。
无示例类增量学习（EF-CIL）是一项挑战性任务，要求在不存储和重放旧类别样本的情况下，持续扩充模型对新类别的识别能力并保留已有知识。现有理论指导框架通过为每个任务训练特定模型来共享网络，将遗忘压力转移到任务标识预测上。但在EF-CIL中，由于缺乏跨任务交互（如样本重放），任务标识预测更为困难。为此，本文从理论上分析了保持特征空间判别性与一致性的必要性和可行性，提出一种名为DCNet的新方法。该方法将类别表示逐步映射至超球面空间，使不同类别正交分布以实现充分的类间分离；同时引入补偿性训练，自适应调节监督强度，从而对齐类内聚合程度。大量实验与理论分析验证了所提方法的优越性。
原文摘要 · Abstract (English)
Exemplar-free class incremental learning (EF-CIL) is a nontrivial task that requires continuously enriching model capability with new classes while maintaining previously learned knowledge without storing and replaying any old class exemplars. An emerging theory-guided framework for CIL trains task-specific models for a shared network, shifting the pressure of forgetting to task-id prediction. In EF-CIL, task-id prediction is more challenging due to the lack of inter-task interaction (e.g., replays of exemplars). To address this issue, we conduct a theoretical analysis of the importance and feasibility of preserving a discriminative and consistent feature space, upon which we propose a novel method termed DCNet. Concretely, it progressively maps class representations into a hyperspherical space, in which different classes are orthogonally distributed to achieve ample inter-class separation. Meanwhile, it also introduces compensatory training to adaptively adjust supervision intensity, thereby aligning the degree of intra-class aggregation. Extensive experiments and theoretical analysis verified the superiority of the proposed DCNet.
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。