We present a novel semi-supervised learning framework that intelligently leverages the consistency regularization between the model's predictions from two strongly-augmented views of an image, weighted by a confidence of pseudo-label, dubbed ConMatch. While the latest semi-supervised learning methods use weakly- and strongly-augmented views of an image to define a directional consistency loss, how to define such direction for the consistency regularization between two strongly-augmented views remains unexplored. To account for this, we present novel confidence measures for pseudo-labels from strongly-augmented views by means of weakly-augmented view as an anchor in non-parametric and parametric approaches. Especially, in parametric approach, we present, for the first time, to learn the confidence of pseudo-label within the networks, which is learned with backbone model in an end-to-end manner. In addition, we also present a stage-wise training to boost the convergence of training. When incorporated in existing semi-supervised learners, ConMatch consistently boosts the performance. We conduct experiments to demonstrate the effectiveness of our ConMatch over the latest methods and provide extensive ablation studies. Code has been made publicly available at https://github.com/JiwonCocoder/ConMatch.
翻译:我们提出了一个新的半监督的学习框架,它明智地利用了模型从两种强烈放大的图像观点对模型预测的一致性进行规范化,这些观点的精美度来自两种强烈放大的图像观点,这些观点通过假标签、假标签、称为ConMatch的自信加以权衡。虽然最新的半监督的学习方法使用微弱和强烈放大的图像观点来界定方向一致性损失,但如何为两种强烈放大的观点之间的一致性规范化确定这种方向仍未得到探讨。为此,我们为来自强烈放大的观点的伪标签提出了新的信任度措施,采用的方法是微弱的放大观点作为非参数和参数方法的锚点。特别是,在对准方法方面,我们首次展示了对网络内伪标签的信任度的薄弱性和强度;我们以端到端的方式学习了主干模型所学的伪标签;此外,我们还介绍了一个分阶段的培训,以促进培训的趋同。当我们被纳入现有的半监督学习者中时,ConMatch不断提升业绩。我们进行了广泛的实验,在对准方法进行广泛的研究时,并展示了Mqow的最新方法。