[论文解读] Learning Adaptive Embedding Considering Incremental Class

Yang Yang, Zhenqiang Sun+4 more|arXiv (Cornell University)|Aug 31, 2020

Domain Adaptation and Few-Shot Learning参考文献 43被引用 9

一句话总结

该论文提出CILF，一种新颖的类增量学习框架，通过解耦的基于原型的损失函数和可学习的课程聚类算子，联合解决新类检测与灾难性遗忘问题。该方法能够准确检测多个未知类，并在最小遗忘下实现有效的模型更新，在流式分类基准上优于现有方法。

ABSTRACT

Class-Incremental Learning (CIL) aims to train a reliable model with the streaming data, which emerges unknown classes sequentially. Different from traditional closed set learning, CIL has two main challenges: 1) Novel class detection. The initial training data only contains incomplete classes, and streaming test data will accept unknown classes. Therefore, the model needs to not only accurately classify known classes, but also effectively detect unknown classes; 2) Model expansion. After the novel classes are detected, the model needs to be updated without re-training using entire previous data. However, traditional CIL methods have not fully considered these two challenges, first, they are always restricted to single novel class detection each phase and embedding confusion caused by unknown classes. Besides, they also ignore the catastrophic forgetting of known categories in model update. To this end, we propose a Class-Incremental Learning without Forgetting (CILF) framework, which aims to learn adaptive embedding for processing novel class detection and model update in a unified framework. In detail, CILF designs to regularize classification with decoupled prototype based loss, which can improve the intra-class and inter-class structure significantly, and acquire a compact embedding representation for novel class detection in result. Then, CILF employs a learnable curriculum clustering operator to estimate the number of semantic clusters via fine-tuning the learned network, in which curriculum operator can adaptively learn the embedding in self-taught form. Therefore, CILF can detect multiple novel classes and mitigate the embedding confusion problem. Last, with the labeled streaming test data, CILF can update the network with robust regularization to mitigate the catastrophic forgetting. Consequently, CILF is able to iteratively perform novel class detection and model update.

研究动机与目标

• 解决类增量学习（CIL）中流数据下的新类检测与灾难性遗忘双重挑战。
• 克服先前方法在假设仅检测单个新类且对嵌入混淆缺乏鲁棒性的局限性。
• 开发一个统一框架，支持无需完整微调的迭代式检测与模型更新。
• 通过记忆蒸馏和原型正则化，在增量模型更新过程中减轻灾难性遗忘。
• 在不依赖先验语义信息的前提下，实现对多个新类的检测，仅依赖学习到的嵌入表示。

提出的方法

• 引入解耦的基于原型的损失函数，以增强类内紧凑性与类间可分性，提升新类检测的嵌入质量。
• 设计一种可学习的课程聚类算子，通过自教自导的方式微调网络，自适应估计语义聚类数量。
• 在增量模型更新过程中应用记忆蒸馏与原型正则化，以保留先前学习类别的知识。
• 在测试阶段采用基于置信度的检测方法，并结合边界损失，以区分已知类与未知类。
• 将课程聚类算子与特征空间集成，以减少嵌入混淆并提升聚类收敛性。
• 采用联合优化目标，通过超参数λ₁和λ₂平衡分类准确率、新类检测与遗忘缓解。

实验结果

研究问题

• RQ1 在流式学习中，统一框架能否在无先验语义信息的情况下有效检测多个新类？
• RQ2 在类增量学习中，如何在增量模型更新过程中减轻已知类别的灾难性遗忘？
• RQ3 课程聚类算子在多大程度上减少了嵌入混淆并提升了检测性能？
• RQ4 所提出的解耦原型损失在嵌入空间中如何增强类内与类间结构？
• RQ5 标注查询规模与超参数设置对检测与遗忘性能有何影响？

主要发现

• CILF在所有评估基准上均实现了最低的遗忘率，已知类别准确率下降速度慢于基线方法。
• 在CIFAR-10数据集上，CILF在新类检测的AUROC、F1与FPR95指标上均优于所有基线方法，尤其在多新类场景下表现更优。
• 课程聚类算子显著减少了嵌入混淆，T-SNE可视化结果表明类间分离更加清晰。
• 由于采用高效的基于课程的聚类策略，CILF执行时间快于竞争方法，避免了因嵌入混淆导致的缓慢收敛。
• 随着标注查询规模增大，性能持续提升，证实了真实标签数据在有效模型更新中的重要性。
• 超参数敏感性分析表明，当λ₁ > 1时性能达到最优，表明强正则化对检测与遗忘缓解至关重要。