[论文解读] Rethinking Re-Sampling in Imbalanced Semi-Supervised Learning

Ju He, Adam Kortylewski+5 more|arXiv (Cornell University)|Jun 1, 2021

Domain Adaptation and Few-Shot Learning参考文献 33被引用 15

一句话总结

本文提出Bi-Sampling（BiS），一种新颖的两阶段重采样策略，用于类别不平衡的半监督学习，通过解耦特征提取器与分类器的训练过程。BiS 逐步从不平衡采样（以保持表征学习）过渡到平衡采样（以提升分类器泛化能力），在 CIFAR-10 和 CIFAR-100 上实现了最先进性能，相较于基线模型最高提升达 7.5%。

ABSTRACT

Semi-Supervised Learning (SSL) has shown its strong ability in utilizing unlabeled data when labeled data is scarce. However, most SSL algorithms work under the assumption that the class distributions are balanced in both training and test sets. In this work, we consider the problem of SSL on class-imbalanced data, which better reflects real-world situations. In particular, we decouple the training of the representation and the classifier, and systematically investigate the effects of different data re-sampling techniques when training the whole network including a classifier as well as fine-tuning the feature extractor only. We find that data re-sampling is of critical importance to learn a good classifier as it increases the accuracy of the pseudo-labels, in particular for the minority classes in the unlabeled data. Interestingly, we find that accurate pseudo-labels do not help when training the feature extractor, rather contrariwise, data re-sampling harms the training of the feature extractor. This finding is against the general intuition that wrong pseudo-labels always harm the model performance in SSL. Based on these findings, we suggest to re-think the current paradigm of having a single data re-sampling strategy and develop a simple yet highly effective Bi-Sampling (BiS) strategy for SSL on class-imbalanced data. BiS implements two different re-sampling strategies for training the feature extractor and the classifier and integrates this decoupled training into an end-to-end framework. In particular, BiS progressively changes the data distribution during training such that in the beginning the feature extractor is trained effectively, while towards the end of the training the data is re-balanced such that the classifier is trained reliably. We benchmark our proposed bi-sampling strategy extensively on popular datasets and achieve state-of-the-art performances.

研究动机与目标

• 研究数据重采样对类别不平衡半监督学习中特征提取器与分类器学习的影响。
• 挑战在整个训练过程中使用单一重采样策略的常规范式。
• 开发一种方法，通过使用不同的采样策略，分别优化表征学习与分类器校准。
• 在如 CIFAR-10 和 CIFAR-100 等类别不平衡半监督基准上实现最先进性能。

提出的方法

• BiS 采用两种不同的重采样策略：使用随机采样器训练特征提取器，使用均值采样器训练分类器。
• 通过采样权重 α 的衰减调度，训练过程中逐步从随机采样器过渡到均值采样器。
• 采样权重 α 沿训练步数按抛物线函数衰减，早期偏向不平衡数据，后期偏向平衡数据。
• 在特征提取器与分类器联合训练后，对分类器进行重采样微调，从而解耦学习过程。
• 保持概率 q 控制对未标注数据重采样的强度，消融实验表明中间值时性能最优。
• 该框架端到端集成，无需额外训练技术如 mixup。

实验结果

研究问题

• RQ1 在类别不平衡的半监督学习中，数据重采样如何影响特征提取器与分类器的学习？
• RQ2 当伪标签偏向头部类别时，重采样是否会损害或促进表征学习？
• RQ3 单一重采样策略能否有效平衡长尾 SSL 中的表征学习与分类器校准？
• RQ4 在训练过程中，不同重采样策略之间的过渡应采用何种最优调度？
• RQ5 采用针对特征提取器与分类器分别采样的解耦训练策略，是否能优于联合重采样？

主要发现

• 尽管伪标签存在偏差，重采样仍会损害特征提取器的训练，表明表征学习从不平衡数据中受益。
• 分类器性能在重采样下显著提升，尤其对少数类而言，即使伪标签不准确也表现优异。
• 在 CIFAR-10 上，BiS 相较于 FixMatch 基线模型，β=2 且 λ=100 时实现 7.5% 的绝对性能提升。
• 在 CIFAR-100 上，BiS 相较于 FixMatch 提升 4%，相较于 CReST+ 提升 1%，在相似设置下表现更优。
• α 的抛物线衰减调度相比线性、余弦和等量衰减，性能提升约 8%。
• 消融研究显示，保持概率 q 的极端值（0 或 1）会降低性能，最优值位于中间范围。