双向优于单向：基于循环一致性的双向对齐用于无样本类增量学习
Two-Way Is Better Than One: Bidirectional Alignment with Cycle Consistency for Exemplar-Free Class-Incremental Learning
AI总结：
提出BiCyc方法，通过双向投影器对齐和循环一致性目标，解决无样本类增量学习中原型漂移和单向投影偏差问题，减少灾难性遗忘并提升准确率。
摘要：
持续学习（CL）旨在使模型在不遗忘先前知识的情况下获取新技能。在无样本类增量学习（EFCIL）中，由于无法存储过去数据，这一挑战被放大，旧类的表示漂移尤其有害。基于原型的EFCIL因其高效性而具有吸引力，但随着嵌入空间的演化，原型会发生漂移；因此，基于投影的漂移补偿已成为一种流行的补救措施。然而，我们表明，现有的单向投影引入了系统性偏差：它们要么追溯性地扭曲当前特征几何结构，要么仅局部对齐旧类，导致跨任务累积的循环不一致性。我们提出BiCyc，一种具有循环一致性目标的双向投影器对齐方法。BiCyc联合优化两个映射（旧到新和新到旧），并采用停止梯度门控，使得传输和表示共同演化。分析表明，循环损失在白化空间中将奇异谱向单位值收缩，并且类均值和协方差的改进传输导致分类对数几率扰动更小，从而保留旧类决策并减轻灾难性遗忘。实验上，在标准EFCIL基准测试中，BiCyc显著减少了遗忘并提高了从头开始设置下的准确率，同时在预训练细粒度场景中保持竞争力。
英文摘要：
Continual learning (CL) seeks models that acquire new skills without erasing prior knowledge. In exemplar-free class-incremental learning (EFCIL), this challenge is amplified because past data cannot be stored, making representation drift for old classes particularly harmful. Prototype-based EFCIL is attractive for its efficiency, yet prototypes drift as the embedding space evolves; therefore, projection-based drift compensation has become a popular remedy. We show, however, that existing one-directional projections introduce systematic bias: they either retroactively distort the current feature geometry or align past classes only locally, leaving cycle inconsistencies that accumulate across tasks. We introduce BiCyc, a bidirectional projector alignment approach with a cycle-consistency objective. BiCyc jointly optimizes two maps, old-to-new and new-to-old, with stop-gradient gating so that transport and representation co-evolve. Analytically, we show that the cycle loss contracts the singular spectrum toward unity in whitened space, and that improved transport of class means and covariances yields smaller perturbations of classification log-odds, preserving old-class decisions and mitigating catastrophic forgetting. Empirically, across standard EFCIL benchmarks, BiCyc substantially reduces forgetting and improves accuracy in from-scratch settings, while remaining competitive in the pretrained fine-grained regime.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
