ContraCluster：通过对比自监督与基于原型的半监督实现无标签分类学习
ContraCluster: Learning to Classify without Labels by Contrastive Self-Supervision and Prototype-Based Semi-Supervision
AI总结：
ContraCluster 提出一种结合对比自监督预训练、对比原型采样和基于原型的半监督微调三阶段的无监督图像分类方法，在 CIFAR-10 等基准上以 90.8% 准确率显著超越现有方法，逼近监督学习性能。
摘要：
表示学习的最新进展激励我们以有原则的方式应对无监督图像分类任务这一具有挑战性的问题。我们提出 ContraCluster，一种将聚类与对比自监督学习能力相结合的无监督图像分类方法。ContraCluster 由三个阶段组成：（1）对比自监督预训练（CPT），（2）对比原型采样（CPS），以及（3）基于原型的半监督微调（PB-SFT）。CPS 能够在对比学习学到的嵌入空间中选取高精度、具有类别代表性的原型图像。我们将采样得到的原型作为带噪标签数据，执行基于原型的半监督微调（PB-SFT），利用少量原型和大量无标签数据进一步提升准确率。我们通过实验证明，ContraCluster 在包括 CIFAR-10、STL-10 和 ImageNet-10 在内的标准基准数据集上取得了新的最先进结果。例如，ContraCluster 在 CIFAR-10 上达到约 90.8% 的准确率，大幅优于 DAC（52.2%）、IIC（61.7%）和 SCAN（87.6%）。在完全没有任何标签的情况下，ContraCluster 可以达到 90.8% 的准确率，与最佳监督对照方法的 95.8% 相当。
英文摘要：
The recent advances in representation learning inspire us to take on the challenging problem of unsupervised image classification tasks in a principled way. We propose ContraCluster, an unsupervised image classification method that combines clustering with the power of contrastive self-supervised learning. ContraCluster consists of three stages: (1) contrastive self-supervised pre-training (CPT), (2) contrastive prototype sampling (CPS), and (3) prototype-based semi-supervised fine-tuning (PB-SFT). CPS can select highly accurate, categorically prototypical images in an embedding space learned by contrastive learning. We use sampled prototypes as noisy labeled data to perform semi-supervised fine-tuning (PB-SFT), leveraging small prototypes and large unlabeled data to further enhance the accuracy. We demonstrate empirically that ContraCluster achieves new state-of-the-art results for standard benchmark datasets including CIFAR-10, STL-10, and ImageNet-10. For example, ContraCluster achieves about 90.8% accuracy for CIFAR-10, which outperforms DAC (52.2%), IIC (61.7%), and SCAN (87.6%) by a large margin. Without any labels, ContraCluster can achieve a 90.8% accuracy that is comparable to 95.8% by the best supervised counterpart.
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。
